id
large_string | submitter
large_string | authors
large_string | title
large_string | comments
large_string | journal-ref
large_string | doi
large_string | report-no
large_string | categories
large_string | license
large_string | abstract
large_string | versions
large list | update_date
timestamp[ms] | authors_parsed
large list | prompt
large_string | label
string | prob
float64 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2205.12718 | Jihang Wang | Jihang Wang, Dongcheng Zhao, Guobin Shen, Qian Zhang, Yi Zeng | DPSNN: A Differentially Private Spiking Neural Network with Temporal
Enhanced Pooling | 24 pages, 8 figures | null | null | null | cs.NE cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Privacy protection is a crucial issue in machine learning algorithms, and the
current privacy protection is combined with traditional artificial neural
networks based on real values. Spiking neural network (SNN), the new generation
of artificial neural networks, plays a crucial role in many fields. Therefore,
research on the privacy protection of SNN is urgently needed. This paper
combines the differential privacy(DP) algorithm with SNN and proposes a
differentially private spiking neural network (DPSNN). The SNN uses discrete
spike sequences to transmit information, combined with the gradient noise
introduced by DP so that SNN maintains strong privacy protection. At the same
time, to make SNN maintain high performance while obtaining high privacy
protection, we propose the temporal enhanced pooling (TEP) method. It fully
integrates the temporal information of SNN into the spatial information
transfer, which enables SNN to perform better information transfer. We conduct
experiments on static and neuromorphic datasets, and the experimental results
show that our algorithm still maintains high performance while providing strong
privacy protection.
| [
{
"version": "v1",
"created": "Tue, 24 May 2022 05:27:53 GMT"
},
{
"version": "v2",
"created": "Thu, 9 Feb 2023 02:28:56 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Apr 2023 11:44:15 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wang",
"Jihang",
""
],
[
"Zhao",
"Dongcheng",
""
],
[
"Shen",
"Guobin",
""
],
[
"Zhang",
"Qian",
""
],
[
"Zeng",
"Yi",
""
]
] | TITLE: DPSNN: A Differentially Private Spiking Neural Network with Temporal
Enhanced Pooling
ABSTRACT: Privacy protection is a crucial issue in machine learning algorithms, and the
current privacy protection is combined with traditional artificial neural
networks based on real values. Spiking neural network (SNN), the new generation
of artificial neural networks, plays a crucial role in many fields. Therefore,
research on the privacy protection of SNN is urgently needed. This paper
combines the differential privacy(DP) algorithm with SNN and proposes a
differentially private spiking neural network (DPSNN). The SNN uses discrete
spike sequences to transmit information, combined with the gradient noise
introduced by DP so that SNN maintains strong privacy protection. At the same
time, to make SNN maintain high performance while obtaining high privacy
protection, we propose the temporal enhanced pooling (TEP) method. It fully
integrates the temporal information of SNN into the spatial information
transfer, which enables SNN to perform better information transfer. We conduct
experiments on static and neuromorphic datasets, and the experimental results
show that our algorithm still maintains high performance while providing strong
privacy protection.
| no_new_dataset | 0.948728 |
2205.14430 | Liang Zhou | Kaiyi Zhang, Liang Zhou, Lu Chen, Shitong He, Daniel Weiskopf, Yunhai
Wang | Angle-Uniform Parallel Coordinates | Computational Visual Media, 2023 | null | 10.1007/s41095-022-0291-7 | null | cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present angle-uniform parallel coordinates, a data-independent technique
that deforms the image plane of parallel coordinates so that the angles of
linear relationships between two variables are linearly mapped along the
horizontal axis of the parallel coordinates plot. Despite being a common method
for visualizing multidimensional data, parallel coordinates are ineffective for
revealing positive correlations since the associated parallel coordinates
points of such structures may be located at infinity in the image plane and the
asymmetric encoding of negative and positive correlations may lead to
unreliable estimations. To address this issue, we introduce a transformation
that bounds all points horizontally using an angle-uniform mapping and shrinks
them vertically in a structure-preserving fashion; polygonal lines become
smooth curves and a symmetric representation of data correlations is achieved.
We further propose a combined subsampling and density visualization approach to
reduce visual clutter caused by overdrawing. Our method enables accurate visual
pattern interpretation of data correlations, and its data-independent nature
makes it applicable to all multidimensional datasets. The usefulness of our
method is demonstrated using examples of synthetic and real-world datasets.
| [
{
"version": "v1",
"created": "Sat, 28 May 2022 13:24:37 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 07:02:06 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Zhang",
"Kaiyi",
""
],
[
"Zhou",
"Liang",
""
],
[
"Chen",
"Lu",
""
],
[
"He",
"Shitong",
""
],
[
"Weiskopf",
"Daniel",
""
],
[
"Wang",
"Yunhai",
""
]
] | TITLE: Angle-Uniform Parallel Coordinates
ABSTRACT: We present angle-uniform parallel coordinates, a data-independent technique
that deforms the image plane of parallel coordinates so that the angles of
linear relationships between two variables are linearly mapped along the
horizontal axis of the parallel coordinates plot. Despite being a common method
for visualizing multidimensional data, parallel coordinates are ineffective for
revealing positive correlations since the associated parallel coordinates
points of such structures may be located at infinity in the image plane and the
asymmetric encoding of negative and positive correlations may lead to
unreliable estimations. To address this issue, we introduce a transformation
that bounds all points horizontally using an angle-uniform mapping and shrinks
them vertically in a structure-preserving fashion; polygonal lines become
smooth curves and a symmetric representation of data correlations is achieved.
We further propose a combined subsampling and density visualization approach to
reduce visual clutter caused by overdrawing. Our method enables accurate visual
pattern interpretation of data correlations, and its data-independent nature
makes it applicable to all multidimensional datasets. The usefulness of our
method is demonstrated using examples of synthetic and real-world datasets.
| no_new_dataset | 0.956594 |
2206.02082 | Xudong Lin | Xudong Lin, Simran Tiwari, Shiyuan Huang, Manling Li, Mike Zheng Shou,
Heng Ji, Shih-Fu Chang | Towards Fast Adaptation of Pretrained Contrastive Models for
Multi-channel Video-Language Retrieval | To appear in CVPR 2023; The code will be released at
https://github.com/XudongLinthu/upgradable-multimodal-intelligence | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-channel video-language retrieval require models to understand
information from different channels (e.g. video$+$question, video$+$speech) to
correctly link a video with a textual response or query. Fortunately,
contrastive multimodal models are shown to be highly effective at aligning
entities in images/videos and text, e.g., CLIP; text contrastive models are
extensively studied recently for their strong ability of producing
discriminative sentence embeddings, e.g., SimCSE. However, there is not a clear
way to quickly adapt these two lines to multi-channel video-language retrieval
with limited data and resources. In this paper, we identify a principled model
design space with two axes: how to represent videos and how to fuse video and
text information. Based on categorization of recent methods, we investigate the
options of representing videos using continuous feature vectors or discrete
text tokens; for the fusion method, we explore the use of a multimodal
transformer or a pretrained contrastive text model. We extensively evaluate the
four combinations on five video-language datasets. We surprisingly find that
discrete text tokens coupled with a pretrained contrastive text model yields
the best performance, which can even outperform state-of-the-art on the iVQA
and How2QA datasets without additional training on millions of video-text data.
Further analysis shows that this is because representing videos as text tokens
captures the key visual information and text tokens are naturally aligned with
text models that are strong retrievers after the contrastive pretraining
process. All the empirical analysis establishes a solid foundation for future
research on affordable and upgradable multimodal intelligence.
| [
{
"version": "v1",
"created": "Sun, 5 Jun 2022 01:43:52 GMT"
},
{
"version": "v2",
"created": "Fri, 6 Jan 2023 14:19:23 GMT"
},
{
"version": "v3",
"created": "Tue, 7 Mar 2023 15:17:57 GMT"
},
{
"version": "v4",
"created": "Tue, 11 Apr 2023 02:29:20 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Lin",
"Xudong",
""
],
[
"Tiwari",
"Simran",
""
],
[
"Huang",
"Shiyuan",
""
],
[
"Li",
"Manling",
""
],
[
"Shou",
"Mike Zheng",
""
],
[
"Ji",
"Heng",
""
],
[
"Chang",
"Shih-Fu",
""
]
] | TITLE: Towards Fast Adaptation of Pretrained Contrastive Models for
Multi-channel Video-Language Retrieval
ABSTRACT: Multi-channel video-language retrieval require models to understand
information from different channels (e.g. video$+$question, video$+$speech) to
correctly link a video with a textual response or query. Fortunately,
contrastive multimodal models are shown to be highly effective at aligning
entities in images/videos and text, e.g., CLIP; text contrastive models are
extensively studied recently for their strong ability of producing
discriminative sentence embeddings, e.g., SimCSE. However, there is not a clear
way to quickly adapt these two lines to multi-channel video-language retrieval
with limited data and resources. In this paper, we identify a principled model
design space with two axes: how to represent videos and how to fuse video and
text information. Based on categorization of recent methods, we investigate the
options of representing videos using continuous feature vectors or discrete
text tokens; for the fusion method, we explore the use of a multimodal
transformer or a pretrained contrastive text model. We extensively evaluate the
four combinations on five video-language datasets. We surprisingly find that
discrete text tokens coupled with a pretrained contrastive text model yields
the best performance, which can even outperform state-of-the-art on the iVQA
and How2QA datasets without additional training on millions of video-text data.
Further analysis shows that this is because representing videos as text tokens
captures the key visual information and text tokens are naturally aligned with
text models that are strong retrievers after the contrastive pretraining
process. All the empirical analysis establishes a solid foundation for future
research on affordable and upgradable multimodal intelligence.
| no_new_dataset | 0.949949 |
2206.14579 | Fenglin Liu | Fenglin Liu, Shen Ge, Yuexian Zou, Xian Wu | Competence-based Multimodal Curriculum Learning for Medical Report
Generation | Accepted by ACL 2021 (Oral) | null | null | null | cs.CL cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Medical report generation task, which targets to produce long and coherent
descriptions of medical images, has attracted growing research interests
recently. Different from the general image captioning tasks, medical report
generation is more challenging for data-driven neural models. This is mainly
due to 1) the serious data bias and 2) the limited medical data. To alleviate
the data bias and make best use of available data, we propose a
Competence-based Multimodal Curriculum Learning framework (CMCL). Specifically,
CMCL simulates the learning process of radiologists and optimizes the model in
a step by step manner. Firstly, CMCL estimates the difficulty of each training
instance and evaluates the competence of current model; Secondly, CMCL selects
the most suitable batch of training instances considering current model
competence. By iterating above two steps, CMCL can gradually improve the
model's performance. The experiments on the public IU-Xray and MIMIC-CXR
datasets show that CMCL can be incorporated into existing models to improve
their performance.
| [
{
"version": "v1",
"created": "Fri, 24 Jun 2022 08:16:01 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Oct 2022 22:46:05 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Apr 2023 06:23:39 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Liu",
"Fenglin",
""
],
[
"Ge",
"Shen",
""
],
[
"Zou",
"Yuexian",
""
],
[
"Wu",
"Xian",
""
]
] | TITLE: Competence-based Multimodal Curriculum Learning for Medical Report
Generation
ABSTRACT: Medical report generation task, which targets to produce long and coherent
descriptions of medical images, has attracted growing research interests
recently. Different from the general image captioning tasks, medical report
generation is more challenging for data-driven neural models. This is mainly
due to 1) the serious data bias and 2) the limited medical data. To alleviate
the data bias and make best use of available data, we propose a
Competence-based Multimodal Curriculum Learning framework (CMCL). Specifically,
CMCL simulates the learning process of radiologists and optimizes the model in
a step by step manner. Firstly, CMCL estimates the difficulty of each training
instance and evaluates the competence of current model; Secondly, CMCL selects
the most suitable batch of training instances considering current model
competence. By iterating above two steps, CMCL can gradually improve the
model's performance. The experiments on the public IU-Xray and MIMIC-CXR
datasets show that CMCL can be incorporated into existing models to improve
their performance.
| no_new_dataset | 0.943867 |
2207.13394 | Aythami Morales | Daniel DeAlcala and Aythami Morales and Ruben Tolosana and Alejandro
Acien and Julian Fierrez and Santiago Hernandez and Miguel A. Ferrer and
Moises Diaz | BeCAPTCHA-Type: Biometric Keystroke Data Generation for Improved Bot
Detection | Paper accepted in IEEE Computer Society Workshop on Biometrics
(CVPRw) 2023 | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This work proposes a data driven learning model for the synthesis of
keystroke biometric data. The proposed method is compared with two statistical
approaches based on Universal and User-dependent models. These approaches are
validated on the bot detection task, using the keystroke synthetic data to
improve the training process of keystroke-based bot detection systems. Our
experimental framework considers a dataset with 136 million keystroke events
from 168 thousand subjects. We have analyzed the performance of the three
synthesis approaches through qualitative and quantitative experiments.
Different bot detectors are considered based on several supervised classifiers
(Support Vector Machine, Random Forest, Gaussian Naive Bayes and a Long
Short-Term Memory network) and a learning framework including human and
synthetic samples. The experiments demonstrate the realism of the synthetic
samples. The classification results suggest that in scenarios with large
labeled data, these synthetic samples can be detected with high accuracy.
However, in few-shot learning scenarios it represents an important challenge.
Furthermore, these results show the great potential of the presented models.
| [
{
"version": "v1",
"created": "Wed, 27 Jul 2022 09:26:15 GMT"
},
{
"version": "v2",
"created": "Thu, 28 Jul 2022 06:59:43 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Apr 2023 15:36:42 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"DeAlcala",
"Daniel",
""
],
[
"Morales",
"Aythami",
""
],
[
"Tolosana",
"Ruben",
""
],
[
"Acien",
"Alejandro",
""
],
[
"Fierrez",
"Julian",
""
],
[
"Hernandez",
"Santiago",
""
],
[
"Ferrer",
"Miguel A.",
""
],
[
"Diaz",
"Moises",
""
]
] | TITLE: BeCAPTCHA-Type: Biometric Keystroke Data Generation for Improved Bot
Detection
ABSTRACT: This work proposes a data driven learning model for the synthesis of
keystroke biometric data. The proposed method is compared with two statistical
approaches based on Universal and User-dependent models. These approaches are
validated on the bot detection task, using the keystroke synthetic data to
improve the training process of keystroke-based bot detection systems. Our
experimental framework considers a dataset with 136 million keystroke events
from 168 thousand subjects. We have analyzed the performance of the three
synthesis approaches through qualitative and quantitative experiments.
Different bot detectors are considered based on several supervised classifiers
(Support Vector Machine, Random Forest, Gaussian Naive Bayes and a Long
Short-Term Memory network) and a learning framework including human and
synthetic samples. The experiments demonstrate the realism of the synthetic
samples. The classification results suggest that in scenarios with large
labeled data, these synthetic samples can be detected with high accuracy.
However, in few-shot learning scenarios it represents an important challenge.
Furthermore, these results show the great potential of the presented models.
| no_new_dataset | 0.916633 |
2208.08579 | Mukund Sudarshan | Mukund Sudarshan, Aahlad Manas Puli, Wesley Tansey, Rajesh Ranganath | DIET: Conditional independence testing with marginal dependence measures
of residual information | null | null | null | null | stat.ME cs.LG stat.ML | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Conditional randomization tests (CRTs) assess whether a variable $x$ is
predictive of another variable $y$, having observed covariates $z$. CRTs
require fitting a large number of predictive models, which is often
computationally intractable. Existing solutions to reduce the cost of CRTs
typically split the dataset into a train and test portion, or rely on
heuristics for interactions, both of which lead to a loss in power. We propose
the decoupled independence test (DIET), an algorithm that avoids both of these
issues by leveraging marginal independence statistics to test conditional
independence relationships. DIET tests the marginal independence of two random
variables: $F(x \mid z)$ and $F(y \mid z)$ where $F(\cdot \mid z)$ is a
conditional cumulative distribution function (CDF). These variables are termed
"information residuals." We give sufficient conditions for DIET to achieve
finite sample type-1 error control and power greater than the type-1 error
rate. We then prove that when using the mutual information between the
information residuals as a test statistic, DIET yields the most powerful
conditionally valid test. Finally, we show DIET achieves higher power than
other tractable CRTs on several synthetic and real benchmarks.
| [
{
"version": "v1",
"created": "Thu, 18 Aug 2022 00:48:04 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 06:48:52 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Sudarshan",
"Mukund",
""
],
[
"Puli",
"Aahlad Manas",
""
],
[
"Tansey",
"Wesley",
""
],
[
"Ranganath",
"Rajesh",
""
]
] | TITLE: DIET: Conditional independence testing with marginal dependence measures
of residual information
ABSTRACT: Conditional randomization tests (CRTs) assess whether a variable $x$ is
predictive of another variable $y$, having observed covariates $z$. CRTs
require fitting a large number of predictive models, which is often
computationally intractable. Existing solutions to reduce the cost of CRTs
typically split the dataset into a train and test portion, or rely on
heuristics for interactions, both of which lead to a loss in power. We propose
the decoupled independence test (DIET), an algorithm that avoids both of these
issues by leveraging marginal independence statistics to test conditional
independence relationships. DIET tests the marginal independence of two random
variables: $F(x \mid z)$ and $F(y \mid z)$ where $F(\cdot \mid z)$ is a
conditional cumulative distribution function (CDF). These variables are termed
"information residuals." We give sufficient conditions for DIET to achieve
finite sample type-1 error control and power greater than the type-1 error
rate. We then prove that when using the mutual information between the
information residuals as a test statistic, DIET yields the most powerful
conditionally valid test. Finally, we show DIET achieves higher power than
other tractable CRTs on several synthetic and real benchmarks.
| no_new_dataset | 0.939248 |
2208.10833 | Hongcheng Guo | Hongcheng Guo, Yuhui Guo, Renjie Chen, Jian Yang, Jiaheng Liu, Zhoujun
Li, Tieqiao Zheng, Weichao Hou, Liangfan Zheng, Bo Zhang | LogLG: Weakly Supervised Log Anomaly Detection via Log-Event Graph
Construction | 12 pages | null | null | null | cs.SE cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fully supervised log anomaly detection methods suffer the heavy burden of
annotating massive unlabeled log data. Recently, many semi-supervised methods
have been proposed to reduce annotation costs with the help of parsed
templates. However, these methods consider each keyword independently, which
disregards the correlation between keywords and the contextual relationships
among log sequences. In this paper, we propose a novel weakly supervised log
anomaly detection framework, named LogLG, to explore the semantic connections
among keywords from sequences. Specifically, we design an end-to-end iterative
process, where the keywords of unlabeled logs are first extracted to construct
a log-event graph. Then, we build a subgraph annotator to generate pseudo
labels for unlabeled log sequences. To ameliorate the annotation quality, we
adopt a self-supervised task to pre-train a subgraph annotator. After that, a
detection model is trained with the generated pseudo labels. Conditioned on the
classification results, we re-extract the keywords from the log sequences and
update the log-event graph for the next iteration. Experiments on five
benchmarks validate the effectiveness of LogLG for detecting anomalies on
unlabeled log data and demonstrate that LogLG, as the state-of-the-art weakly
supervised method, achieves significant performance improvements compared to
existing methods.
| [
{
"version": "v1",
"created": "Tue, 23 Aug 2022 09:32:19 GMT"
},
{
"version": "v2",
"created": "Thu, 25 Aug 2022 06:42:18 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Sep 2022 02:36:58 GMT"
},
{
"version": "v4",
"created": "Thu, 2 Feb 2023 09:17:52 GMT"
},
{
"version": "v5",
"created": "Tue, 11 Apr 2023 07:46:32 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Guo",
"Hongcheng",
""
],
[
"Guo",
"Yuhui",
""
],
[
"Chen",
"Renjie",
""
],
[
"Yang",
"Jian",
""
],
[
"Liu",
"Jiaheng",
""
],
[
"Li",
"Zhoujun",
""
],
[
"Zheng",
"Tieqiao",
""
],
[
"Hou",
"Weichao",
""
],
[
"Zheng",
"Liangfan",
""
],
[
"Zhang",
"Bo",
""
]
] | TITLE: LogLG: Weakly Supervised Log Anomaly Detection via Log-Event Graph
Construction
ABSTRACT: Fully supervised log anomaly detection methods suffer the heavy burden of
annotating massive unlabeled log data. Recently, many semi-supervised methods
have been proposed to reduce annotation costs with the help of parsed
templates. However, these methods consider each keyword independently, which
disregards the correlation between keywords and the contextual relationships
among log sequences. In this paper, we propose a novel weakly supervised log
anomaly detection framework, named LogLG, to explore the semantic connections
among keywords from sequences. Specifically, we design an end-to-end iterative
process, where the keywords of unlabeled logs are first extracted to construct
a log-event graph. Then, we build a subgraph annotator to generate pseudo
labels for unlabeled log sequences. To ameliorate the annotation quality, we
adopt a self-supervised task to pre-train a subgraph annotator. After that, a
detection model is trained with the generated pseudo labels. Conditioned on the
classification results, we re-extract the keywords from the log sequences and
update the log-event graph for the next iteration. Experiments on five
benchmarks validate the effectiveness of LogLG for detecting anomalies on
unlabeled log data and demonstrate that LogLG, as the state-of-the-art weakly
supervised method, achieves significant performance improvements compared to
existing methods.
| no_new_dataset | 0.9462 |
2208.11266 | Xiaofan Yu | Xiaofan Yu, Yunhui Guo, Sicun Gao, Tajana Rosing | SCALE: Online Self-Supervised Lifelong Learning without Prior Knowledge | Accepted by CLVision'23 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Unsupervised lifelong learning refers to the ability to learn over time while
memorizing previous patterns without supervision. Although great progress has
been made in this direction, existing work often assumes strong prior knowledge
about the incoming data (e.g., knowing the class boundaries), which can be
impossible to obtain in complex and unpredictable environments. In this paper,
motivated by real-world scenarios, we propose a more practical problem setting
called online self-supervised lifelong learning without prior knowledge. The
proposed setting is challenging due to the non-iid and single-pass data, the
absence of external supervision, and no prior knowledge. To address the
challenges, we propose Self-Supervised ContrAstive Lifelong LEarning without
Prior Knowledge (SCALE) which can extract and memorize representations on the
fly purely from the data continuum. SCALE is designed around three major
components: a pseudo-supervised contrastive loss, a self-supervised forgetting
loss, and an online memory update for uniform subset selection. All three
components are designed to work collaboratively to maximize learning
performance. We perform comprehensive experiments of SCALE under iid and four
non-iid data streams. The results show that SCALE outperforms the
state-of-the-art algorithm in all settings with improvements up to 3.83%, 2.77%
and 5.86% in terms of kNN accuracy on CIFAR-10, CIFAR-100, and TinyImageNet
datasets.
| [
{
"version": "v1",
"created": "Wed, 24 Aug 2022 02:01:06 GMT"
},
{
"version": "v2",
"created": "Mon, 26 Sep 2022 00:15:05 GMT"
},
{
"version": "v3",
"created": "Mon, 3 Oct 2022 21:42:42 GMT"
},
{
"version": "v4",
"created": "Wed, 23 Nov 2022 09:20:15 GMT"
},
{
"version": "v5",
"created": "Mon, 10 Apr 2023 21:25:24 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Yu",
"Xiaofan",
""
],
[
"Guo",
"Yunhui",
""
],
[
"Gao",
"Sicun",
""
],
[
"Rosing",
"Tajana",
""
]
] | TITLE: SCALE: Online Self-Supervised Lifelong Learning without Prior Knowledge
ABSTRACT: Unsupervised lifelong learning refers to the ability to learn over time while
memorizing previous patterns without supervision. Although great progress has
been made in this direction, existing work often assumes strong prior knowledge
about the incoming data (e.g., knowing the class boundaries), which can be
impossible to obtain in complex and unpredictable environments. In this paper,
motivated by real-world scenarios, we propose a more practical problem setting
called online self-supervised lifelong learning without prior knowledge. The
proposed setting is challenging due to the non-iid and single-pass data, the
absence of external supervision, and no prior knowledge. To address the
challenges, we propose Self-Supervised ContrAstive Lifelong LEarning without
Prior Knowledge (SCALE) which can extract and memorize representations on the
fly purely from the data continuum. SCALE is designed around three major
components: a pseudo-supervised contrastive loss, a self-supervised forgetting
loss, and an online memory update for uniform subset selection. All three
components are designed to work collaboratively to maximize learning
performance. We perform comprehensive experiments of SCALE under iid and four
non-iid data streams. The results show that SCALE outperforms the
state-of-the-art algorithm in all settings with improvements up to 3.83%, 2.77%
and 5.86% in terms of kNN accuracy on CIFAR-10, CIFAR-100, and TinyImageNet
datasets.
| no_new_dataset | 0.94801 |
2209.03683 | Juan Ozaita Corral | Miguel Ru\'iz-Garc\'ia, Juan Ozaita, Mar\'ia Pereda, Antonio Alfonso,
Pablo Bra\~nas-Garza, Jose A. Cuesta and \'Angel S\'anchez | Known by the company we keep: `Triadic influence' as a proxy for
compatibility in social relationships | null | Proceedings of the National Academy of Sciences (USA), 120,
e2215041120 (2023) | 10.1073/pnas.2215041120 | null | cs.SI cond-mat.dis-nn stat.ML | http://creativecommons.org/licenses/by/4.0/ | Networks of social interactions are the substrate upon which civilizations
are built. Often, we create new bonds with people that we like or feel that our
relationships are damaged through the intervention of third parties. Despite
their importance and the huge impact that these processes have in our lives,
quantitative scientific understanding of them is still in its infancy, mainly
due to the difficulty of collecting large datasets of social networks including
individual attributes. In this work, we present a thorough study of real social
networks of 13 schools, with more than 3,000 students and 60,000 declared
positive and negative relations, including tests for personal traits of all the
students. We introduce a metric -- the `triadic influence' -- that measures the
influence of nearest-neighbors in the relationships of their contacts. We use
neural networks to predict the relationships and to extract the probability
that two students are friends or enemies depending on their personal attributes
or the triadic influence. We alternatively use a high-dimensional embedding of
the network structure to also predict the relationships. Remarkably, the
triadic influence (a simple one-dimensional metric) achieves the highest
accuracy at predicting the relationship between two students. We postulate that
the probabilities extracted from the neural networks -- functions of the
triadic influence and the personalities of the students -- control the
evolution of real social networks, opening a new avenue for the quantitative
study of these systems.
| [
{
"version": "v1",
"created": "Thu, 8 Sep 2022 09:52:34 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Sep 2022 08:53:05 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Ruíz-García",
"Miguel",
""
],
[
"Ozaita",
"Juan",
""
],
[
"Pereda",
"María",
""
],
[
"Alfonso",
"Antonio",
""
],
[
"Brañas-Garza",
"Pablo",
""
],
[
"Cuesta",
"Jose A.",
""
],
[
"Sánchez",
"Ángel",
""
]
] | TITLE: Known by the company we keep: `Triadic influence' as a proxy for
compatibility in social relationships
ABSTRACT: Networks of social interactions are the substrate upon which civilizations
are built. Often, we create new bonds with people that we like or feel that our
relationships are damaged through the intervention of third parties. Despite
their importance and the huge impact that these processes have in our lives,
quantitative scientific understanding of them is still in its infancy, mainly
due to the difficulty of collecting large datasets of social networks including
individual attributes. In this work, we present a thorough study of real social
networks of 13 schools, with more than 3,000 students and 60,000 declared
positive and negative relations, including tests for personal traits of all the
students. We introduce a metric -- the `triadic influence' -- that measures the
influence of nearest-neighbors in the relationships of their contacts. We use
neural networks to predict the relationships and to extract the probability
that two students are friends or enemies depending on their personal attributes
or the triadic influence. We alternatively use a high-dimensional embedding of
the network structure to also predict the relationships. Remarkably, the
triadic influence (a simple one-dimensional metric) achieves the highest
accuracy at predicting the relationship between two students. We postulate that
the probabilities extracted from the neural networks -- functions of the
triadic influence and the personalities of the students -- control the
evolution of real social networks, opening a new avenue for the quantitative
study of these systems.
| no_new_dataset | 0.932576 |
2209.15182 | Ruiqi Wang | Ruiqi Wang, Wonse Jo, Dezhong Zhao, Weizheng Wang, Baijian Yang,
Guohua Chen and Byung-Cheol Min | Husformer: A Multi-Modal Transformer for Multi-Modal Human State
Recognition | null | null | null | null | cs.HC cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Human state recognition is a critical topic with pervasive and important
applications in human-machine systems. Multi-modal fusion, the combination of
metrics from multiple data sources, has been shown as a sound method for
improving the recognition performance. However, while promising results have
been reported by recent multi-modal-based models, they generally fail to
leverage the sophisticated fusion strategies that would model sufficient
cross-modal interactions when producing the fusion representation; instead,
current methods rely on lengthy and inconsistent data preprocessing and feature
crafting. To address this limitation, we propose an end-to-end multi-modal
transformer framework for multi-modal human state recognition called Husformer.
Specifically, we propose to use cross-modal transformers, which inspire one
modality to reinforce itself through directly attending to latent relevance
revealed in other modalities, to fuse different modalities while ensuring
sufficient awareness of the cross-modal interactions introduced. Subsequently,
we utilize a self-attention transformer to further prioritize contextual
information in the fusion representation. Using two such attention mechanisms
enables effective and adaptive adjustments to noise and interruptions in
multi-modal signals during the fusion process and in relation to high-level
features. Extensive experiments on two human emotion corpora (DEAP and WESAD)
and two cognitive workload datasets (MOCAS and CogLoad) demonstrate that in the
recognition of human state, our Husformer outperforms both state-of-the-art
multi-modal baselines and the use of a single modality by a large margin,
especially when dealing with raw multi-modal signals. We also conducted an
ablation study to show the benefits of each component in Husformer.
| [
{
"version": "v1",
"created": "Fri, 30 Sep 2022 02:11:27 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 03:48:45 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wang",
"Ruiqi",
""
],
[
"Jo",
"Wonse",
""
],
[
"Zhao",
"Dezhong",
""
],
[
"Wang",
"Weizheng",
""
],
[
"Yang",
"Baijian",
""
],
[
"Chen",
"Guohua",
""
],
[
"Min",
"Byung-Cheol",
""
]
] | TITLE: Husformer: A Multi-Modal Transformer for Multi-Modal Human State
Recognition
ABSTRACT: Human state recognition is a critical topic with pervasive and important
applications in human-machine systems. Multi-modal fusion, the combination of
metrics from multiple data sources, has been shown as a sound method for
improving the recognition performance. However, while promising results have
been reported by recent multi-modal-based models, they generally fail to
leverage the sophisticated fusion strategies that would model sufficient
cross-modal interactions when producing the fusion representation; instead,
current methods rely on lengthy and inconsistent data preprocessing and feature
crafting. To address this limitation, we propose an end-to-end multi-modal
transformer framework for multi-modal human state recognition called Husformer.
Specifically, we propose to use cross-modal transformers, which inspire one
modality to reinforce itself through directly attending to latent relevance
revealed in other modalities, to fuse different modalities while ensuring
sufficient awareness of the cross-modal interactions introduced. Subsequently,
we utilize a self-attention transformer to further prioritize contextual
information in the fusion representation. Using two such attention mechanisms
enables effective and adaptive adjustments to noise and interruptions in
multi-modal signals during the fusion process and in relation to high-level
features. Extensive experiments on two human emotion corpora (DEAP and WESAD)
and two cognitive workload datasets (MOCAS and CogLoad) demonstrate that in the
recognition of human state, our Husformer outperforms both state-of-the-art
multi-modal baselines and the use of a single modality by a large margin,
especially when dealing with raw multi-modal signals. We also conducted an
ablation study to show the benefits of each component in Husformer.
| no_new_dataset | 0.9463 |
2210.00092 | Karan Singhal | Raviteja Vemulapalli, Warren Richard Morningstar, Philip Andrew
Mansfield, Hubert Eichner, Karan Singhal, Arash Afkanpour, Bradley Green | Federated Training of Dual Encoding Models on Small Non-IID Client
Datasets | ICLR 2023 Workshop on Pitfalls of Limited Data and Computation for
Trustworthy ML | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Dual encoding models that encode a pair of inputs are widely used for
representation learning. Many approaches train dual encoding models by
maximizing agreement between pairs of encodings on centralized training data.
However, in many scenarios, datasets are inherently decentralized across many
clients (user devices or organizations) due to privacy concerns, motivating
federated learning. In this work, we focus on federated training of dual
encoding models on decentralized data composed of many small, non-IID
(independent and identically distributed) client datasets. We show that
existing approaches that work well in centralized settings perform poorly when
naively adapted to this setting using federated averaging. We observe that, we
can simulate large-batch loss computation on individual clients for loss
functions that are based on encoding statistics. Based on this insight, we
propose a novel federated training approach, Distributed Cross Correlation
Optimization (DCCO), which trains dual encoding models using encoding
statistics aggregated across clients, without sharing individual data samples.
Our experimental results on two datasets demonstrate that the proposed DCCO
approach outperforms federated variants of existing approaches by a large
margin.
| [
{
"version": "v1",
"created": "Fri, 30 Sep 2022 21:03:43 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Apr 2023 18:27:06 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Vemulapalli",
"Raviteja",
""
],
[
"Morningstar",
"Warren Richard",
""
],
[
"Mansfield",
"Philip Andrew",
""
],
[
"Eichner",
"Hubert",
""
],
[
"Singhal",
"Karan",
""
],
[
"Afkanpour",
"Arash",
""
],
[
"Green",
"Bradley",
""
]
] | TITLE: Federated Training of Dual Encoding Models on Small Non-IID Client
Datasets
ABSTRACT: Dual encoding models that encode a pair of inputs are widely used for
representation learning. Many approaches train dual encoding models by
maximizing agreement between pairs of encodings on centralized training data.
However, in many scenarios, datasets are inherently decentralized across many
clients (user devices or organizations) due to privacy concerns, motivating
federated learning. In this work, we focus on federated training of dual
encoding models on decentralized data composed of many small, non-IID
(independent and identically distributed) client datasets. We show that
existing approaches that work well in centralized settings perform poorly when
naively adapted to this setting using federated averaging. We observe that, we
can simulate large-batch loss computation on individual clients for loss
functions that are based on encoding statistics. Based on this insight, we
propose a novel federated training approach, Distributed Cross Correlation
Optimization (DCCO), which trains dual encoding models using encoding
statistics aggregated across clients, without sharing individual data samples.
Our experimental results on two datasets demonstrate that the proposed DCCO
approach outperforms federated variants of existing approaches by a large
margin.
| no_new_dataset | 0.947137 |
2210.10914 | Fenglin Liu | Fenglin Liu, Xuancheng Ren, Xian Wu, Wei Fan, Yuexian Zou, Xu Sun | Prophet Attention: Predicting Attention with Future Attention for Image
Captioning | Accepted by NeurIPS 2020 | null | null | null | cs.CV cs.CL | http://creativecommons.org/licenses/by/4.0/ | Recently, attention based models have been used extensively in many
sequence-to-sequence learning systems. Especially for image captioning, the
attention based models are expected to ground correct image regions with proper
generated words. However, for each time step in the decoding process, the
attention based models usually use the hidden state of the current input to
attend to the image regions. Under this setting, these attention models have a
"deviated focus" problem that they calculate the attention weights based on
previous words instead of the one to be generated, impairing the performance of
both grounding and captioning. In this paper, we propose the Prophet Attention,
similar to the form of self-supervision. In the training stage, this module
utilizes the future information to calculate the "ideal" attention weights
towards image regions. These calculated "ideal" weights are further used to
regularize the "deviated" attention. In this manner, image regions are grounded
with the correct words. The proposed Prophet Attention can be easily
incorporated into existing image captioning models to improve their performance
of both grounding and captioning. The experiments on the Flickr30k Entities and
the MSCOCO datasets show that the proposed Prophet Attention consistently
outperforms baselines in both automatic metrics and human evaluations. It is
worth noticing that we set new state-of-the-arts on the two benchmark datasets
and achieve the 1st place on the leaderboard of the online MSCOCO benchmark in
terms of the default ranking score, i.e., CIDEr-c40.
| [
{
"version": "v1",
"created": "Wed, 19 Oct 2022 22:29:31 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 06:12:01 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Liu",
"Fenglin",
""
],
[
"Ren",
"Xuancheng",
""
],
[
"Wu",
"Xian",
""
],
[
"Fan",
"Wei",
""
],
[
"Zou",
"Yuexian",
""
],
[
"Sun",
"Xu",
""
]
] | TITLE: Prophet Attention: Predicting Attention with Future Attention for Image
Captioning
ABSTRACT: Recently, attention based models have been used extensively in many
sequence-to-sequence learning systems. Especially for image captioning, the
attention based models are expected to ground correct image regions with proper
generated words. However, for each time step in the decoding process, the
attention based models usually use the hidden state of the current input to
attend to the image regions. Under this setting, these attention models have a
"deviated focus" problem that they calculate the attention weights based on
previous words instead of the one to be generated, impairing the performance of
both grounding and captioning. In this paper, we propose the Prophet Attention,
similar to the form of self-supervision. In the training stage, this module
utilizes the future information to calculate the "ideal" attention weights
towards image regions. These calculated "ideal" weights are further used to
regularize the "deviated" attention. In this manner, image regions are grounded
with the correct words. The proposed Prophet Attention can be easily
incorporated into existing image captioning models to improve their performance
of both grounding and captioning. The experiments on the Flickr30k Entities and
the MSCOCO datasets show that the proposed Prophet Attention consistently
outperforms baselines in both automatic metrics and human evaluations. It is
worth noticing that we set new state-of-the-arts on the two benchmark datasets
and achieve the 1st place on the leaderboard of the online MSCOCO benchmark in
terms of the default ranking score, i.e., CIDEr-c40.
| no_new_dataset | 0.948251 |
2210.11075 | Haotian Ye | Haotian Ye, James Zou, Linjun Zhang | Freeze then Train: Towards Provable Representation Learning under
Spurious Correlations and Feature Noise | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The existence of spurious correlations such as image backgrounds in the
training environment can make empirical risk minimization (ERM) perform badly
in the test environment. To address this problem, Kirichenko et al. (2022)
empirically found that the core features that are related to the outcome can
still be learned well even with the presence of spurious correlations. This
opens a promising strategy to first train a feature learner rather than a
classifier, and then perform linear probing (last layer retraining) in the test
environment. However, a theoretical understanding of when and why this approach
works is lacking. In this paper, we find that core features are only learned
well when their associated non-realizable noise is smaller than that of
spurious features, which is not necessarily true in practice. We provide both
theories and experiments to support this finding and to illustrate the
importance of non-realizable noise. Moreover, we propose an algorithm called
Freeze then Train (FTT), that first freezes certain salient features and then
trains the rest of the features using ERM. We theoretically show that FTT
preserves features that are more beneficial to test time probing. Across two
commonly used spurious correlation datasets, FTT outperforms ERM, IRM, JTT and
CVaR-DRO, with substantial improvement in accuracy (by 4.5%) when the feature
noise is large. FTT also performs better on general distribution shift
benchmarks.
| [
{
"version": "v1",
"created": "Thu, 20 Oct 2022 07:59:26 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 09:05:45 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Ye",
"Haotian",
""
],
[
"Zou",
"James",
""
],
[
"Zhang",
"Linjun",
""
]
] | TITLE: Freeze then Train: Towards Provable Representation Learning under
Spurious Correlations and Feature Noise
ABSTRACT: The existence of spurious correlations such as image backgrounds in the
training environment can make empirical risk minimization (ERM) perform badly
in the test environment. To address this problem, Kirichenko et al. (2022)
empirically found that the core features that are related to the outcome can
still be learned well even with the presence of spurious correlations. This
opens a promising strategy to first train a feature learner rather than a
classifier, and then perform linear probing (last layer retraining) in the test
environment. However, a theoretical understanding of when and why this approach
works is lacking. In this paper, we find that core features are only learned
well when their associated non-realizable noise is smaller than that of
spurious features, which is not necessarily true in practice. We provide both
theories and experiments to support this finding and to illustrate the
importance of non-realizable noise. Moreover, we propose an algorithm called
Freeze then Train (FTT), that first freezes certain salient features and then
trains the rest of the features using ERM. We theoretically show that FTT
preserves features that are more beneficial to test time probing. Across two
commonly used spurious correlation datasets, FTT outperforms ERM, IRM, JTT and
CVaR-DRO, with substantial improvement in accuracy (by 4.5%) when the feature
noise is large. FTT also performs better on general distribution shift
benchmarks.
| no_new_dataset | 0.945801 |
2210.13634 | Alberto Tono | Alberto Tono and Heyaojing Huang and Ashwin Agrawal and Martin Fischer | Vitruvio: 3D Building Meshes via Single Perspective Sketches | null | null | null | null | cs.CV cs.AI cs.GR cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Today's architectural engineering and construction (AEC) software require a
learning curve to generate a three-dimension building representation. This
limits the ability to quickly validate the volumetric implications of an
initial design idea communicated via a single sketch. Allowing designers to
translate a single sketch to a 3D building will enable owners to instantly
visualize 3D project information without the cognitive load required. If
previous state-of-the-art (SOTA) data-driven methods for single view
reconstruction (SVR) showed outstanding results in the reconstruction process
from a single image or sketch, they lacked specific applications, analysis, and
experiments in the AEC. Therefore, this research addresses this gap,
introducing the first deep learning method focused only on buildings that aim
to convert a single sketch to a 3D building mesh: Vitruvio. Vitruvio adapts
Occupancy Network for SVR tasks on a specific building dataset (Manhattan 1K).
This adaptation brings two main improvements. First, it accelerates the
inference process by more than 26% (from 0.5s to 0.37s). Second, it increases
the reconstruction accuracy (measured by the Chamfer Distance) by 18%. During
this adaptation in the AEC domain, we evaluate the effect of the building
orientation in the learning procedure since it constitutes an important design
factor. While aligning all the buildings to a canonical pose improved the
overall quantitative metrics, it did not capture fine-grain details in more
complex building shapes (as shown in our qualitative analysis). Finally,
Vitruvio outputs a 3D-printable building mesh with arbitrary topology and genus
from a single perspective sketch, providing a step forward to allow owners and
designers to communicate 3D information via a 2D, effective, intuitive, and
universal communication medium: the sketch.
| [
{
"version": "v1",
"created": "Mon, 24 Oct 2022 22:24:58 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 16:52:01 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Tono",
"Alberto",
""
],
[
"Huang",
"Heyaojing",
""
],
[
"Agrawal",
"Ashwin",
""
],
[
"Fischer",
"Martin",
""
]
] | TITLE: Vitruvio: 3D Building Meshes via Single Perspective Sketches
ABSTRACT: Today's architectural engineering and construction (AEC) software require a
learning curve to generate a three-dimension building representation. This
limits the ability to quickly validate the volumetric implications of an
initial design idea communicated via a single sketch. Allowing designers to
translate a single sketch to a 3D building will enable owners to instantly
visualize 3D project information without the cognitive load required. If
previous state-of-the-art (SOTA) data-driven methods for single view
reconstruction (SVR) showed outstanding results in the reconstruction process
from a single image or sketch, they lacked specific applications, analysis, and
experiments in the AEC. Therefore, this research addresses this gap,
introducing the first deep learning method focused only on buildings that aim
to convert a single sketch to a 3D building mesh: Vitruvio. Vitruvio adapts
Occupancy Network for SVR tasks on a specific building dataset (Manhattan 1K).
This adaptation brings two main improvements. First, it accelerates the
inference process by more than 26% (from 0.5s to 0.37s). Second, it increases
the reconstruction accuracy (measured by the Chamfer Distance) by 18%. During
this adaptation in the AEC domain, we evaluate the effect of the building
orientation in the learning procedure since it constitutes an important design
factor. While aligning all the buildings to a canonical pose improved the
overall quantitative metrics, it did not capture fine-grain details in more
complex building shapes (as shown in our qualitative analysis). Finally,
Vitruvio outputs a 3D-printable building mesh with arbitrary topology and genus
from a single perspective sketch, providing a step forward to allow owners and
designers to communicate 3D information via a 2D, effective, intuitive, and
universal communication medium: the sketch.
| no_new_dataset | 0.949669 |
2211.00024 | Davide Piras | Davide Piras, Hiranya V. Peiris, Andrew Pontzen, Luisa Lucie-Smith,
Ningyuan Guo, Brian Nord | A robust estimator of mutual information for deep learning
interpretability | 30 pages, 8 figures. Minor changes to match version accepted for
publication in Machine Learning: Science and Technology. GMM-MI available at
https://github.com/dpiras/GMM-MI | Machine Learning: Science and Technology, Volume 4, Number 2,
025006, April 2023 | 10.1088/2632-2153/acc444 | null | physics.data-an astro-ph.IM cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We develop the use of mutual information (MI), a well-established metric in
information theory, to interpret the inner workings of deep learning models. To
accurately estimate MI from a finite number of samples, we present GMM-MI
(pronounced $``$Jimmie$"$), an algorithm based on Gaussian mixture models that
can be applied to both discrete and continuous settings. GMM-MI is
computationally efficient, robust to the choice of hyperparameters and provides
the uncertainty on the MI estimate due to the finite sample size. We
extensively validate GMM-MI on toy data for which the ground truth MI is known,
comparing its performance against established mutual information estimators. We
then demonstrate the use of our MI estimator in the context of representation
learning, working with synthetic data and physical datasets describing highly
non-linear processes. We train deep learning models to encode high-dimensional
data within a meaningful compressed (latent) representation, and use GMM-MI to
quantify both the level of disentanglement between the latent variables, and
their association with relevant physical quantities, thus unlocking the
interpretability of the latent representation. We make GMM-MI publicly
available.
| [
{
"version": "v1",
"created": "Mon, 31 Oct 2022 18:00:02 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 16:18:11 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Piras",
"Davide",
""
],
[
"Peiris",
"Hiranya V.",
""
],
[
"Pontzen",
"Andrew",
""
],
[
"Lucie-Smith",
"Luisa",
""
],
[
"Guo",
"Ningyuan",
""
],
[
"Nord",
"Brian",
""
]
] | TITLE: A robust estimator of mutual information for deep learning
interpretability
ABSTRACT: We develop the use of mutual information (MI), a well-established metric in
information theory, to interpret the inner workings of deep learning models. To
accurately estimate MI from a finite number of samples, we present GMM-MI
(pronounced $``$Jimmie$"$), an algorithm based on Gaussian mixture models that
can be applied to both discrete and continuous settings. GMM-MI is
computationally efficient, robust to the choice of hyperparameters and provides
the uncertainty on the MI estimate due to the finite sample size. We
extensively validate GMM-MI on toy data for which the ground truth MI is known,
comparing its performance against established mutual information estimators. We
then demonstrate the use of our MI estimator in the context of representation
learning, working with synthetic data and physical datasets describing highly
non-linear processes. We train deep learning models to encode high-dimensional
data within a meaningful compressed (latent) representation, and use GMM-MI to
quantify both the level of disentanglement between the latent variables, and
their association with relevant physical quantities, thus unlocking the
interpretability of the latent representation. We make GMM-MI publicly
available.
| no_new_dataset | 0.942876 |
2211.02247 | Junghyun (Tony) Koo | Junghyun Koo, Marco A. Mart\'inez-Ram\'irez, Wei-Hsiang Liao, Stefan
Uhlich, Kyogu Lee, Yuki Mitsufuji | Music Mixing Style Transfer: A Contrastive Learning Approach to
Disentangle Audio Effects | null | null | null | null | eess.AS cs.LG cs.SD | http://creativecommons.org/licenses/by/4.0/ | We propose an end-to-end music mixing style transfer system that converts the
mixing style of an input multitrack to that of a reference song. This is
achieved with an encoder pre-trained with a contrastive objective to extract
only audio effects related information from a reference music recording. All
our models are trained in a self-supervised manner from an already-processed
wet multitrack dataset with an effective data preprocessing method that
alleviates the data scarcity of obtaining unprocessed dry data. We analyze the
proposed encoder for the disentanglement capability of audio effects and also
validate its performance for mixing style transfer through both objective and
subjective evaluations. From the results, we show the proposed system not only
converts the mixing style of multitrack audio close to a reference but is also
robust with mixture-wise style transfer upon using a music source separation
model.
| [
{
"version": "v1",
"created": "Fri, 4 Nov 2022 03:45:17 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Mar 2023 05:24:20 GMT"
},
{
"version": "v3",
"created": "Tue, 11 Apr 2023 07:53:14 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Koo",
"Junghyun",
""
],
[
"Martínez-Ramírez",
"Marco A.",
""
],
[
"Liao",
"Wei-Hsiang",
""
],
[
"Uhlich",
"Stefan",
""
],
[
"Lee",
"Kyogu",
""
],
[
"Mitsufuji",
"Yuki",
""
]
] | TITLE: Music Mixing Style Transfer: A Contrastive Learning Approach to
Disentangle Audio Effects
ABSTRACT: We propose an end-to-end music mixing style transfer system that converts the
mixing style of an input multitrack to that of a reference song. This is
achieved with an encoder pre-trained with a contrastive objective to extract
only audio effects related information from a reference music recording. All
our models are trained in a self-supervised manner from an already-processed
wet multitrack dataset with an effective data preprocessing method that
alleviates the data scarcity of obtaining unprocessed dry data. We analyze the
proposed encoder for the disentanglement capability of audio effects and also
validate its performance for mixing style transfer through both objective and
subjective evaluations. From the results, we show the proposed system not only
converts the mixing style of multitrack audio close to a reference but is also
robust with mixture-wise style transfer upon using a music source separation
model.
| no_new_dataset | 0.947332 |
2211.07881 | Hengrui Zhang | Hengrui Zhang, Wei Wayne Chen, James M. Rondinelli, Wei Chen | ET-AL: Entropy-Targeted Active Learning for Bias Mitigation in Materials
Data | 35 pages, 13 figures, under review | Applied Physics Reviews 10, 021403 (2023) | 10.1063/5.0138913 | null | cond-mat.mtrl-sci cs.DB cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Growing materials data and data-driven informatics drastically promote the
discovery and design of materials. While there are significant advancements in
data-driven models, the quality of data resources is less studied despite its
huge impact on model performance. In this work, we focus on data bias arising
from uneven coverage of materials families in existing knowledge. Observing
different diversities among crystal systems in common materials databases, we
propose an information entropy-based metric for measuring this bias. To
mitigate the bias, we develop an entropy-targeted active learning (ET-AL)
framework, which guides the acquisition of new data to improve the diversity of
underrepresented crystal systems. We demonstrate the capability of ET-AL for
bias mitigation and the resulting improvement in downstream machine learning
models. This approach is broadly applicable to data-driven materials discovery,
including autonomous data acquisition and dataset trimming to reduce bias, as
well as data-driven informatics in other scientific domains.
| [
{
"version": "v1",
"created": "Tue, 15 Nov 2022 04:12:00 GMT"
},
{
"version": "v2",
"created": "Wed, 16 Nov 2022 22:23:21 GMT"
},
{
"version": "v3",
"created": "Wed, 7 Dec 2022 18:28:18 GMT"
},
{
"version": "v4",
"created": "Mon, 20 Feb 2023 04:00:55 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Zhang",
"Hengrui",
""
],
[
"Chen",
"Wei Wayne",
""
],
[
"Rondinelli",
"James M.",
""
],
[
"Chen",
"Wei",
""
]
] | TITLE: ET-AL: Entropy-Targeted Active Learning for Bias Mitigation in Materials
Data
ABSTRACT: Growing materials data and data-driven informatics drastically promote the
discovery and design of materials. While there are significant advancements in
data-driven models, the quality of data resources is less studied despite its
huge impact on model performance. In this work, we focus on data bias arising
from uneven coverage of materials families in existing knowledge. Observing
different diversities among crystal systems in common materials databases, we
propose an information entropy-based metric for measuring this bias. To
mitigate the bias, we develop an entropy-targeted active learning (ET-AL)
framework, which guides the acquisition of new data to improve the diversity of
underrepresented crystal systems. We demonstrate the capability of ET-AL for
bias mitigation and the resulting improvement in downstream machine learning
models. This approach is broadly applicable to data-driven materials discovery,
including autonomous data acquisition and dataset trimming to reduce bias, as
well as data-driven informatics in other scientific domains.
| no_new_dataset | 0.949248 |
2211.08591 | Jaimie Greasley | Jaimie Greasley and Patrick Hosein | Exploring Supervised Machine Learning for Multi-Phase Identification and
Quantification from Powder X-Ray Diffraction Spectra | null | null | 10.1007/s10853-023-08343-4 | null | cond-mat.mtrl-sci cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Powder X-ray diffraction analysis is a critical component of materials
characterization methodologies. Discerning characteristic Bragg intensity peaks
and assigning them to known crystalline phases is the first qualitative step of
evaluating diffraction spectra. Subsequent to phase identification, Rietveld
refinement may be employed to extract the abundance of quantitative,
material-specific parameters hidden within powder data. These characterization
procedures are yet time-consuming and inhibit efficiency in materials science
workflows. The ever-increasing popularity and propulsion of data science
techniques has provided an obvious solution on the course towards materials
analysis automation. Deep learning has become a prime focus for predicting
crystallographic parameters and features from X-ray spectra. However, the
infeasibility of curating large, well-labelled experimental datasets means that
one must resort to a large number of theoretic simulations for powder data
augmentation to effectively train deep models. Herein, we are interested in
conventional supervised learning algorithms in lieu of deep learning for
multi-label crystalline phase identification and quantitative phase analysis
for a biomedical application. First, models were trained using very limited
experimental data. Further, we incorporated simulated XRD data to assess model
generalizability as well as the efficacy of simulation-based training for
predictive analysis in a real-world X-ray diffraction application.
| [
{
"version": "v1",
"created": "Wed, 16 Nov 2022 00:36:13 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Greasley",
"Jaimie",
""
],
[
"Hosein",
"Patrick",
""
]
] | TITLE: Exploring Supervised Machine Learning for Multi-Phase Identification and
Quantification from Powder X-Ray Diffraction Spectra
ABSTRACT: Powder X-ray diffraction analysis is a critical component of materials
characterization methodologies. Discerning characteristic Bragg intensity peaks
and assigning them to known crystalline phases is the first qualitative step of
evaluating diffraction spectra. Subsequent to phase identification, Rietveld
refinement may be employed to extract the abundance of quantitative,
material-specific parameters hidden within powder data. These characterization
procedures are yet time-consuming and inhibit efficiency in materials science
workflows. The ever-increasing popularity and propulsion of data science
techniques has provided an obvious solution on the course towards materials
analysis automation. Deep learning has become a prime focus for predicting
crystallographic parameters and features from X-ray spectra. However, the
infeasibility of curating large, well-labelled experimental datasets means that
one must resort to a large number of theoretic simulations for powder data
augmentation to effectively train deep models. Herein, we are interested in
conventional supervised learning algorithms in lieu of deep learning for
multi-label crystalline phase identification and quantitative phase analysis
for a biomedical application. First, models were trained using very limited
experimental data. Further, we incorporated simulated XRD data to assess model
generalizability as well as the efficacy of simulation-based training for
predictive analysis in a real-world X-ray diffraction application.
| no_new_dataset | 0.949716 |
2211.08703 | Yongjie Chen | Yongjie Chen, Tieru Wu | SATVSR: Scenario Adaptive Transformer for Cross Scenarios Video
Super-Resolution | null | null | 10.1088/1742-6596/2456/1/012028 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Video Super-Resolution (VSR) aims to recover sequences of high-resolution
(HR) frames from low-resolution (LR) frames. Previous methods mainly utilize
temporally adjacent frames to assist the reconstruction of target frames.
However, in the real world, there is a lot of irrelevant information in
adjacent frames of videos with fast scene switching, these VSR methods cannot
adaptively distinguish and select useful information. In contrast, with a
transformer structure suitable for temporal tasks, we devise a novel adaptive
scenario video super-resolution method. Specifically, we use optical flow to
label the patches in each video frame, only calculate the attention of patches
with the same label. Then select the most relevant label among them to
supplement the spatial-temporal information of the target frame. This design
can directly make the supplementary information come from the same scene as
much as possible. We further propose a cross-scale feature aggregation module
to better handle the scale variation problem. Compared with other video
super-resolution methods, our method not only achieves significant performance
gains on single-scene videos but also has better robustness on cross-scene
datasets.
| [
{
"version": "v1",
"created": "Wed, 16 Nov 2022 06:30:13 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Chen",
"Yongjie",
""
],
[
"Wu",
"Tieru",
""
]
] | TITLE: SATVSR: Scenario Adaptive Transformer for Cross Scenarios Video
Super-Resolution
ABSTRACT: Video Super-Resolution (VSR) aims to recover sequences of high-resolution
(HR) frames from low-resolution (LR) frames. Previous methods mainly utilize
temporally adjacent frames to assist the reconstruction of target frames.
However, in the real world, there is a lot of irrelevant information in
adjacent frames of videos with fast scene switching, these VSR methods cannot
adaptively distinguish and select useful information. In contrast, with a
transformer structure suitable for temporal tasks, we devise a novel adaptive
scenario video super-resolution method. Specifically, we use optical flow to
label the patches in each video frame, only calculate the attention of patches
with the same label. Then select the most relevant label among them to
supplement the spatial-temporal information of the target frame. This design
can directly make the supplementary information come from the same scene as
much as possible. We further propose a cross-scale feature aggregation module
to better handle the scale variation problem. Compared with other video
super-resolution methods, our method not only achieves significant performance
gains on single-scene videos but also has better robustness on cross-scene
datasets.
| no_new_dataset | 0.949435 |
2211.12817 | Xiao Liu | Xiao Liu, Ankur Sikarwar, Gabriel Kreiman, Zenglin Shi, Mengmi Zhang | Reason from Context with Self-supervised Learning | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Self-supervised learning (SSL) learns to capture discriminative visual
features useful for knowledge transfers. To better accommodate the
object-centric nature of current downstream tasks such as object recognition
and detection, various methods have been proposed to suppress contextual biases
or disentangle objects from contexts. Nevertheless, these methods may prove
inadequate in situations where object identity needs to be reasoned from
associated context, such as recognizing or inferring tiny or obscured objects.
As an initial effort in the SSL literature, we investigate whether and how
contextual associations can be enhanced for visual reasoning within SSL
regimes, by (a) proposing a new Self-supervised method with external memories
for Context Reasoning (SeCo), and (b) introducing two new downstream tasks,
lift-the-flap and object priming, addressing the problems of "what" and "where"
in context reasoning. In both tasks, SeCo outperformed all state-of-the-art
(SOTA) SSL methods by a significant margin. Our network analysis revealed that
the proposed external memory in SeCo learns to store prior contextual
knowledge, facilitating target identity inference in the lift-the-flap task.
Moreover, we conducted psychophysics experiments and introduced a Human
benchmark in Object Priming dataset (HOP). Our results demonstrate that SeCo
exhibits human-like behaviors.
| [
{
"version": "v1",
"created": "Wed, 23 Nov 2022 10:02:05 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 07:17:38 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Liu",
"Xiao",
""
],
[
"Sikarwar",
"Ankur",
""
],
[
"Kreiman",
"Gabriel",
""
],
[
"Shi",
"Zenglin",
""
],
[
"Zhang",
"Mengmi",
""
]
] | TITLE: Reason from Context with Self-supervised Learning
ABSTRACT: Self-supervised learning (SSL) learns to capture discriminative visual
features useful for knowledge transfers. To better accommodate the
object-centric nature of current downstream tasks such as object recognition
and detection, various methods have been proposed to suppress contextual biases
or disentangle objects from contexts. Nevertheless, these methods may prove
inadequate in situations where object identity needs to be reasoned from
associated context, such as recognizing or inferring tiny or obscured objects.
As an initial effort in the SSL literature, we investigate whether and how
contextual associations can be enhanced for visual reasoning within SSL
regimes, by (a) proposing a new Self-supervised method with external memories
for Context Reasoning (SeCo), and (b) introducing two new downstream tasks,
lift-the-flap and object priming, addressing the problems of "what" and "where"
in context reasoning. In both tasks, SeCo outperformed all state-of-the-art
(SOTA) SSL methods by a significant margin. Our network analysis revealed that
the proposed external memory in SeCo learns to store prior contextual
knowledge, facilitating target identity inference in the lift-the-flap task.
Moreover, we conducted psychophysics experiments and introduced a Human
benchmark in Object Priming dataset (HOP). Our results demonstrate that SeCo
exhibits human-like behaviors.
| new_dataset | 0.965185 |
2212.13899 | Ha Thanh Nguyen | Ha-Thanh Nguyen, Manh-Kien Phi, Xuan-Bach Ngo, Vu Tran, Le-Minh
Nguyen, Minh-Phuong Tu | Attentive Deep Neural Networks for Legal Document Retrieval | Preprint version. The official version will be published in
Artificial Intelligence and Law journal | null | 10.1007/s10506-022-09341-8 | null | cs.IR cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Legal text retrieval serves as a key component in a wide range of legal text
processing tasks such as legal question answering, legal case entailment, and
statute law retrieval. The performance of legal text retrieval depends, to a
large extent, on the representation of text, both query and legal documents.
Based on good representations, a legal text retrieval model can effectively
match the query to its relevant documents. Because legal documents often
contain long articles and only some parts are relevant to queries, it is quite
a challenge for existing models to represent such documents. In this paper, we
study the use of attentive neural network-based text representation for statute
law document retrieval. We propose a general approach using deep neural
networks with attention mechanisms. Based on it, we develop two hierarchical
architectures with sparse attention to represent long sentences and articles,
and we name them Attentive CNN and Paraformer. The methods are evaluated on
datasets of different sizes and characteristics in English, Japanese, and
Vietnamese. Experimental results show that: i) Attentive neural methods
substantially outperform non-neural methods in terms of retrieval performance
across datasets and languages; ii) Pretrained transformer-based models achieve
better accuracy on small datasets at the cost of high computational complexity
while lighter weight Attentive CNN achieves better accuracy on large datasets;
and iii) Our proposed Paraformer outperforms state-of-the-art methods on COLIEE
dataset, achieving the highest recall and F2 scores in the top-N retrieval
task.
| [
{
"version": "v1",
"created": "Tue, 13 Dec 2022 01:37:27 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Nguyen",
"Ha-Thanh",
""
],
[
"Phi",
"Manh-Kien",
""
],
[
"Ngo",
"Xuan-Bach",
""
],
[
"Tran",
"Vu",
""
],
[
"Nguyen",
"Le-Minh",
""
],
[
"Tu",
"Minh-Phuong",
""
]
] | TITLE: Attentive Deep Neural Networks for Legal Document Retrieval
ABSTRACT: Legal text retrieval serves as a key component in a wide range of legal text
processing tasks such as legal question answering, legal case entailment, and
statute law retrieval. The performance of legal text retrieval depends, to a
large extent, on the representation of text, both query and legal documents.
Based on good representations, a legal text retrieval model can effectively
match the query to its relevant documents. Because legal documents often
contain long articles and only some parts are relevant to queries, it is quite
a challenge for existing models to represent such documents. In this paper, we
study the use of attentive neural network-based text representation for statute
law document retrieval. We propose a general approach using deep neural
networks with attention mechanisms. Based on it, we develop two hierarchical
architectures with sparse attention to represent long sentences and articles,
and we name them Attentive CNN and Paraformer. The methods are evaluated on
datasets of different sizes and characteristics in English, Japanese, and
Vietnamese. Experimental results show that: i) Attentive neural methods
substantially outperform non-neural methods in terms of retrieval performance
across datasets and languages; ii) Pretrained transformer-based models achieve
better accuracy on small datasets at the cost of high computational complexity
while lighter weight Attentive CNN achieves better accuracy on large datasets;
and iii) Our proposed Paraformer outperforms state-of-the-art methods on COLIEE
dataset, achieving the highest recall and F2 scores in the top-N retrieval
task.
| no_new_dataset | 0.949902 |
2212.14337 | Christopher Wolters | Christopher Wolters, Brady Taylor, Edward Hanson, Xiaoxuan Yang, Ulf
Schlichtmann and Yiran Chen | Biologically Plausible Learning on Neuromorphic Hardware Architectures | null | null | null | null | cs.NE cs.ET | http://creativecommons.org/licenses/by/4.0/ | With an ever-growing number of parameters defining increasingly complex
networks, Deep Learning has led to several breakthroughs surpassing human
performance. As a result, data movement for these millions of model parameters
causes a growing imbalance known as the memory wall. Neuromorphic computing is
an emerging paradigm that confronts this imbalance by performing computations
directly in analog memories. On the software side, the sequential
Backpropagation algorithm prevents efficient parallelization and thus fast
convergence. A novel method, Direct Feedback Alignment, resolves inherent layer
dependencies by directly passing the error from the output to each layer. At
the intersection of hardware/software co-design, there is a demand for
developing algorithms that are tolerable to hardware nonidealities. Therefore,
this work explores the interrelationship of implementing bio-plausible learning
in-situ on neuromorphic hardware, emphasizing energy, area, and latency
constraints. Using the benchmarking framework DNN+NeuroSim, we investigate the
impact of hardware nonidealities and quantization on algorithm performance, as
well as how network topologies and algorithm-level design choices can scale
latency, energy and area consumption of a chip. To the best of our knowledge,
this work is the first to compare the impact of different learning algorithms
on Compute-In-Memory-based hardware and vice versa. The best results achieved
for accuracy remain Backpropagation-based, notably when facing hardware
imperfections. Direct Feedback Alignment, on the other hand, allows for
significant speedup due to parallelization, reducing training time by a factor
approaching N for N-layered networks.
| [
{
"version": "v1",
"created": "Thu, 29 Dec 2022 15:10:59 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 10:50:10 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wolters",
"Christopher",
""
],
[
"Taylor",
"Brady",
""
],
[
"Hanson",
"Edward",
""
],
[
"Yang",
"Xiaoxuan",
""
],
[
"Schlichtmann",
"Ulf",
""
],
[
"Chen",
"Yiran",
""
]
] | TITLE: Biologically Plausible Learning on Neuromorphic Hardware Architectures
ABSTRACT: With an ever-growing number of parameters defining increasingly complex
networks, Deep Learning has led to several breakthroughs surpassing human
performance. As a result, data movement for these millions of model parameters
causes a growing imbalance known as the memory wall. Neuromorphic computing is
an emerging paradigm that confronts this imbalance by performing computations
directly in analog memories. On the software side, the sequential
Backpropagation algorithm prevents efficient parallelization and thus fast
convergence. A novel method, Direct Feedback Alignment, resolves inherent layer
dependencies by directly passing the error from the output to each layer. At
the intersection of hardware/software co-design, there is a demand for
developing algorithms that are tolerable to hardware nonidealities. Therefore,
this work explores the interrelationship of implementing bio-plausible learning
in-situ on neuromorphic hardware, emphasizing energy, area, and latency
constraints. Using the benchmarking framework DNN+NeuroSim, we investigate the
impact of hardware nonidealities and quantization on algorithm performance, as
well as how network topologies and algorithm-level design choices can scale
latency, energy and area consumption of a chip. To the best of our knowledge,
this work is the first to compare the impact of different learning algorithms
on Compute-In-Memory-based hardware and vice versa. The best results achieved
for accuracy remain Backpropagation-based, notably when facing hardware
imperfections. Direct Feedback Alignment, on the other hand, allows for
significant speedup due to parallelization, reducing training time by a factor
approaching N for N-layered networks.
| no_new_dataset | 0.947914 |
2301.06197 | Hussein Mozannar | Hussein Mozannar, Hunter Lang, Dennis Wei, Prasanna Sattigeri, Subhro
Das, David Sontag | Who Should Predict? Exact Algorithms For Learning to Defer to Humans | AISTATS 2023 | null | null | null | cs.LG cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Automated AI classifiers should be able to defer the prediction to a human
decision maker to ensure more accurate predictions. In this work, we jointly
train a classifier with a rejector, which decides on each data point whether
the classifier or the human should predict. We show that prior approaches can
fail to find a human-AI system with low misclassification error even when there
exists a linear classifier and rejector that have zero error (the realizable
setting). We prove that obtaining a linear pair with low error is NP-hard even
when the problem is realizable. To complement this negative result, we give a
mixed-integer-linear-programming (MILP) formulation that can optimally solve
the problem in the linear setting. However, the MILP only scales to
moderately-sized problems. Therefore, we provide a novel surrogate loss
function that is realizable-consistent and performs well empirically. We test
our approaches on a comprehensive set of datasets and compare to a wide range
of baselines.
| [
{
"version": "v1",
"created": "Sun, 15 Jan 2023 21:57:36 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 07:40:40 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Mozannar",
"Hussein",
""
],
[
"Lang",
"Hunter",
""
],
[
"Wei",
"Dennis",
""
],
[
"Sattigeri",
"Prasanna",
""
],
[
"Das",
"Subhro",
""
],
[
"Sontag",
"David",
""
]
] | TITLE: Who Should Predict? Exact Algorithms For Learning to Defer to Humans
ABSTRACT: Automated AI classifiers should be able to defer the prediction to a human
decision maker to ensure more accurate predictions. In this work, we jointly
train a classifier with a rejector, which decides on each data point whether
the classifier or the human should predict. We show that prior approaches can
fail to find a human-AI system with low misclassification error even when there
exists a linear classifier and rejector that have zero error (the realizable
setting). We prove that obtaining a linear pair with low error is NP-hard even
when the problem is realizable. To complement this negative result, we give a
mixed-integer-linear-programming (MILP) formulation that can optimally solve
the problem in the linear setting. However, the MILP only scales to
moderately-sized problems. Therefore, we provide a novel surrogate loss
function that is realizable-consistent and performs well empirically. We test
our approaches on a comprehensive set of datasets and compare to a wide range
of baselines.
| no_new_dataset | 0.946498 |
2301.06855 | Manasi Muglikar Ms. | Manasi Muglikar, Leonard Bauersfeld, Diederik Paul Moeys, Davide
Scaramuzza | Event-based Shape from Polarization | IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
Vancouver, 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | State-of-the-art solutions for Shape-from-Polarization (SfP) suffer from a
speed-resolution tradeoff: they either sacrifice the number of polarization
angles measured or necessitate lengthy acquisition times due to framerate
constraints, thus compromising either accuracy or latency. We tackle this
tradeoff using event cameras. Event cameras operate at microseconds resolution
with negligible motion blur, and output a continuous stream of events that
precisely measures how light changes over time asynchronously. We propose a
setup that consists of a linear polarizer rotating at high-speeds in front of
an event camera. Our method uses the continuous event stream caused by the
rotation to reconstruct relative intensities at multiple polarizer angles.
Experiments demonstrate that our method outperforms physics-based baselines
using frames, reducing the MAE by 25% in synthetic and real-world dataset. In
the real world, we observe, however, that the challenging conditions (i.e.,
when few events are generated) harm the performance of physics-based solutions.
To overcome this, we propose a learning-based approach that learns to estimate
surface normals even at low event-rates, improving the physics-based approach
by 52% on the real world dataset. The proposed system achieves an acquisition
speed equivalent to 50 fps (>twice the framerate of the commercial polarization
sensor) while retaining the spatial resolution of 1MP. Our evaluation is based
on the first large-scale dataset for event-based SfP
| [
{
"version": "v1",
"created": "Tue, 17 Jan 2023 12:59:58 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 14:50:04 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Muglikar",
"Manasi",
""
],
[
"Bauersfeld",
"Leonard",
""
],
[
"Moeys",
"Diederik Paul",
""
],
[
"Scaramuzza",
"Davide",
""
]
] | TITLE: Event-based Shape from Polarization
ABSTRACT: State-of-the-art solutions for Shape-from-Polarization (SfP) suffer from a
speed-resolution tradeoff: they either sacrifice the number of polarization
angles measured or necessitate lengthy acquisition times due to framerate
constraints, thus compromising either accuracy or latency. We tackle this
tradeoff using event cameras. Event cameras operate at microseconds resolution
with negligible motion blur, and output a continuous stream of events that
precisely measures how light changes over time asynchronously. We propose a
setup that consists of a linear polarizer rotating at high-speeds in front of
an event camera. Our method uses the continuous event stream caused by the
rotation to reconstruct relative intensities at multiple polarizer angles.
Experiments demonstrate that our method outperforms physics-based baselines
using frames, reducing the MAE by 25% in synthetic and real-world dataset. In
the real world, we observe, however, that the challenging conditions (i.e.,
when few events are generated) harm the performance of physics-based solutions.
To overcome this, we propose a learning-based approach that learns to estimate
surface normals even at low event-rates, improving the physics-based approach
by 52% on the real world dataset. The proposed system achieves an acquisition
speed equivalent to 50 fps (>twice the framerate of the commercial polarization
sensor) while retaining the spatial resolution of 1MP. Our evaluation is based
on the first large-scale dataset for event-based SfP
| no_new_dataset | 0.953319 |
2301.07525 | Tong Wu | Tong Wu, Jiarui Zhang, Xiao Fu, Yuxin Wang, Jiawei Ren, Liang Pan,
Wayne Wu, Lei Yang, Jiaqi Wang, Chen Qian, Dahua Lin, Ziwei Liu | OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic
Perception, Reconstruction and Generation | Project page: https://omniobject3d.github.io/ | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in modeling 3D objects mostly rely on synthetic datasets due
to the lack of large-scale realscanned 3D databases. To facilitate the
development of 3D perception, reconstruction, and generation in the real world,
we propose OmniObject3D, a large vocabulary 3D object dataset with massive
high-quality real-scanned 3D objects. OmniObject3D has several appealing
properties: 1) Large Vocabulary: It comprises 6,000 scanned objects in 190
daily categories, sharing common classes with popular 2D datasets (e.g.,
ImageNet and LVIS), benefiting the pursuit of generalizable 3D representations.
2) Rich Annotations: Each 3D object is captured with both 2D and 3D sensors,
providing textured meshes, point clouds, multiview rendered images, and
multiple real-captured videos. 3) Realistic Scans: The professional scanners
support highquality object scans with precise shapes and realistic appearances.
With the vast exploration space offered by OmniObject3D, we carefully set up
four evaluation tracks: a) robust 3D perception, b) novel-view synthesis, c)
neural surface reconstruction, and d) 3D object generation. Extensive studies
are performed on these four benchmarks, revealing new observations, challenges,
and opportunities for future research in realistic 3D vision.
| [
{
"version": "v1",
"created": "Wed, 18 Jan 2023 18:14:18 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 17:41:17 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wu",
"Tong",
""
],
[
"Zhang",
"Jiarui",
""
],
[
"Fu",
"Xiao",
""
],
[
"Wang",
"Yuxin",
""
],
[
"Ren",
"Jiawei",
""
],
[
"Pan",
"Liang",
""
],
[
"Wu",
"Wayne",
""
],
[
"Yang",
"Lei",
""
],
[
"Wang",
"Jiaqi",
""
],
[
"Qian",
"Chen",
""
],
[
"Lin",
"Dahua",
""
],
[
"Liu",
"Ziwei",
""
]
] | TITLE: OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic
Perception, Reconstruction and Generation
ABSTRACT: Recent advances in modeling 3D objects mostly rely on synthetic datasets due
to the lack of large-scale realscanned 3D databases. To facilitate the
development of 3D perception, reconstruction, and generation in the real world,
we propose OmniObject3D, a large vocabulary 3D object dataset with massive
high-quality real-scanned 3D objects. OmniObject3D has several appealing
properties: 1) Large Vocabulary: It comprises 6,000 scanned objects in 190
daily categories, sharing common classes with popular 2D datasets (e.g.,
ImageNet and LVIS), benefiting the pursuit of generalizable 3D representations.
2) Rich Annotations: Each 3D object is captured with both 2D and 3D sensors,
providing textured meshes, point clouds, multiview rendered images, and
multiple real-captured videos. 3) Realistic Scans: The professional scanners
support highquality object scans with precise shapes and realistic appearances.
With the vast exploration space offered by OmniObject3D, we carefully set up
four evaluation tracks: a) robust 3D perception, b) novel-view synthesis, c)
neural surface reconstruction, and d) 3D object generation. Extensive studies
are performed on these four benchmarks, revealing new observations, challenges,
and opportunities for future research in realistic 3D vision.
| new_dataset | 0.964722 |
2301.11219 | Shivam Sharma | Shivam Sharma, Atharva Kulkarni, Tharun Suresh, Himanshi Mathur,
Preslav Nakov, Md. Shad Akhtar, Tanmoy Chakraborty | Characterizing the Entities in Harmful Memes: Who is the Hero, the
Villain, the Victim? | Accepted at EACL 2023 (Main Track). 9 Pages (main content),
Limitations, Ethical Considerations + 4 Pages (Refs.) + Appendix; 8 Figures;
5 Tables; Paper ID: 804 | null | null | null | cs.CL cs.CY | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Memes can sway people's opinions over social media as they combine visual and
textual information in an easy-to-consume manner. Since memes instantly turn
viral, it becomes crucial to infer their intent and potentially associated
harmfulness to take timely measures as needed. A common problem associated with
meme comprehension lies in detecting the entities referenced and characterizing
the role of each of these entities. Here, we aim to understand whether the meme
glorifies, vilifies, or victimizes each entity it refers to. To this end, we
address the task of role identification of entities in harmful memes, i.e.,
detecting who is the 'hero', the 'villain', and the 'victim' in the meme, if
any. We utilize HVVMemes - a memes dataset on US Politics and Covid-19 memes,
released recently as part of the CONSTRAINT@ACL-2022 shared-task. It contains
memes, entities referenced, and their associated roles: hero, villain, victim,
and other. We further design VECTOR (Visual-semantic role dEteCToR), a robust
multi-modal framework for the task, which integrates entity-based contextual
information in the multi-modal representation and compare it to several
standard unimodal (text-only or image-only) or multi-modal (image+text) models.
Our experimental results show that our proposed model achieves an improvement
of 4% over the best baseline and 1% over the best competing stand-alone
submission from the shared-task. Besides divulging an extensive experimental
setup with comparative analyses, we finally highlight the challenges
encountered in addressing the complex task of semantic role labeling within
memes.
| [
{
"version": "v1",
"created": "Thu, 26 Jan 2023 16:55:15 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Apr 2023 20:07:38 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Sharma",
"Shivam",
""
],
[
"Kulkarni",
"Atharva",
""
],
[
"Suresh",
"Tharun",
""
],
[
"Mathur",
"Himanshi",
""
],
[
"Nakov",
"Preslav",
""
],
[
"Akhtar",
"Md. Shad",
""
],
[
"Chakraborty",
"Tanmoy",
""
]
] | TITLE: Characterizing the Entities in Harmful Memes: Who is the Hero, the
Villain, the Victim?
ABSTRACT: Memes can sway people's opinions over social media as they combine visual and
textual information in an easy-to-consume manner. Since memes instantly turn
viral, it becomes crucial to infer their intent and potentially associated
harmfulness to take timely measures as needed. A common problem associated with
meme comprehension lies in detecting the entities referenced and characterizing
the role of each of these entities. Here, we aim to understand whether the meme
glorifies, vilifies, or victimizes each entity it refers to. To this end, we
address the task of role identification of entities in harmful memes, i.e.,
detecting who is the 'hero', the 'villain', and the 'victim' in the meme, if
any. We utilize HVVMemes - a memes dataset on US Politics and Covid-19 memes,
released recently as part of the CONSTRAINT@ACL-2022 shared-task. It contains
memes, entities referenced, and their associated roles: hero, villain, victim,
and other. We further design VECTOR (Visual-semantic role dEteCToR), a robust
multi-modal framework for the task, which integrates entity-based contextual
information in the multi-modal representation and compare it to several
standard unimodal (text-only or image-only) or multi-modal (image+text) models.
Our experimental results show that our proposed model achieves an improvement
of 4% over the best baseline and 1% over the best competing stand-alone
submission from the shared-task. Besides divulging an extensive experimental
setup with comparative analyses, we finally highlight the challenges
encountered in addressing the complex task of semantic role labeling within
memes.
| no_new_dataset | 0.938294 |
2301.11429 | J\"urgen Pfeffer | Juergen Pfeffer, Daniel Matter, Kokil Jaidka, Onur Varol, Afra
Mashhadi, Jana Lasser, Dennis Assenmacher, Siqi Wu, Diyi Yang, Cornelia
Brantner, Daniel M. Romero, Jahna Otterbacher, Carsten Schwemmer, Kenneth
Joseph, David Garcia, Fred Morstatter | Just Another Day on Twitter: A Complete 24 Hours of Twitter Data | null | null | null | null | cs.SI cs.CY | http://creativecommons.org/licenses/by/4.0/ | At the end of October 2022, Elon Musk concluded his acquisition of Twitter.
In the weeks and months before that, several questions were publicly discussed
that were not only of interest to the platform's future buyers, but also of
high relevance to the Computational Social Science research community. For
example, how many active users does the platform have? What percentage of
accounts on the site are bots? And, what are the dominating topics and
sub-topical spheres on the platform? In a globally coordinated effort of 80
scholars to shed light on these questions, and to offer a dataset that will
equip other researchers to do the same, we have collected all 375 million
tweets published within a 24-hour time period starting on September 21, 2022.
To the best of our knowledge, this is the first complete 24-hour Twitter
dataset that is available for the research community. With it, the present work
aims to accomplish two goals. First, we seek to answer the aforementioned
questions and provide descriptive metrics about Twitter that can serve as
references for other researchers. Second, we create a baseline dataset for
future research that can be used to study the potential impact of the
platform's ownership change.
| [
{
"version": "v1",
"created": "Thu, 26 Jan 2023 21:28:40 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 07:33:19 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Pfeffer",
"Juergen",
""
],
[
"Matter",
"Daniel",
""
],
[
"Jaidka",
"Kokil",
""
],
[
"Varol",
"Onur",
""
],
[
"Mashhadi",
"Afra",
""
],
[
"Lasser",
"Jana",
""
],
[
"Assenmacher",
"Dennis",
""
],
[
"Wu",
"Siqi",
""
],
[
"Yang",
"Diyi",
""
],
[
"Brantner",
"Cornelia",
""
],
[
"Romero",
"Daniel M.",
""
],
[
"Otterbacher",
"Jahna",
""
],
[
"Schwemmer",
"Carsten",
""
],
[
"Joseph",
"Kenneth",
""
],
[
"Garcia",
"David",
""
],
[
"Morstatter",
"Fred",
""
]
] | TITLE: Just Another Day on Twitter: A Complete 24 Hours of Twitter Data
ABSTRACT: At the end of October 2022, Elon Musk concluded his acquisition of Twitter.
In the weeks and months before that, several questions were publicly discussed
that were not only of interest to the platform's future buyers, but also of
high relevance to the Computational Social Science research community. For
example, how many active users does the platform have? What percentage of
accounts on the site are bots? And, what are the dominating topics and
sub-topical spheres on the platform? In a globally coordinated effort of 80
scholars to shed light on these questions, and to offer a dataset that will
equip other researchers to do the same, we have collected all 375 million
tweets published within a 24-hour time period starting on September 21, 2022.
To the best of our knowledge, this is the first complete 24-hour Twitter
dataset that is available for the research community. With it, the present work
aims to accomplish two goals. First, we seek to answer the aforementioned
questions and provide descriptive metrics about Twitter that can serve as
references for other researchers. Second, we create a baseline dataset for
future research that can be used to study the potential impact of the
platform's ownership change.
| new_dataset | 0.95388 |
2301.12382 | Maolin Yang | Maolin Yang, Pingyu Jiang, Tianshuo Zang, Yuhao Liu | Data-driven intelligent computational design for products: Method,
techniques, and applications | null | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data-driven intelligent computational design (DICD) is a research hotspot
emerged under the context of fast-developing artificial intelligence. It
emphasizes on utilizing deep learning algorithms to extract and represent the
design features hidden in historical or fabricated design process data, and
then learn the combination and mapping patterns of these design features for
the purposes of design solution retrieval, generation, optimization,
evaluation, etc. Due to its capability of automatically and efficiently
generating design solutions and thus supporting human-in-the-loop intelligent
and innovative design activities, DICD has drawn the attentions from both
academic and industrial fields. However, as an emerging research subject, there
are still many unexplored issues that limit the development and application of
DICD, such as specific dataset building, engineering design related feature
engineering, systematic methods and techniques for DICD implementation in the
entire product design process, etc. In this regard, a systematic and operable
road map for DICD implementation from full-process perspective is established,
including a general workflow for DICD project planning, an overall framework
for DICD project implementation, the computing mechanisms for DICD
implementation, key enabling technologies for detailed DICD implementation, and
three application scenarios of DICD. The road map reveals the common mechanisms
and calculation principles of existing DICD researches, and thus it can provide
systematic guidance for the possible DICD applications that have not been
explored.
| [
{
"version": "v1",
"created": "Sun, 29 Jan 2023 07:17:46 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 07:19:31 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Yang",
"Maolin",
""
],
[
"Jiang",
"Pingyu",
""
],
[
"Zang",
"Tianshuo",
""
],
[
"Liu",
"Yuhao",
""
]
] | TITLE: Data-driven intelligent computational design for products: Method,
techniques, and applications
ABSTRACT: Data-driven intelligent computational design (DICD) is a research hotspot
emerged under the context of fast-developing artificial intelligence. It
emphasizes on utilizing deep learning algorithms to extract and represent the
design features hidden in historical or fabricated design process data, and
then learn the combination and mapping patterns of these design features for
the purposes of design solution retrieval, generation, optimization,
evaluation, etc. Due to its capability of automatically and efficiently
generating design solutions and thus supporting human-in-the-loop intelligent
and innovative design activities, DICD has drawn the attentions from both
academic and industrial fields. However, as an emerging research subject, there
are still many unexplored issues that limit the development and application of
DICD, such as specific dataset building, engineering design related feature
engineering, systematic methods and techniques for DICD implementation in the
entire product design process, etc. In this regard, a systematic and operable
road map for DICD implementation from full-process perspective is established,
including a general workflow for DICD project planning, an overall framework
for DICD project implementation, the computing mechanisms for DICD
implementation, key enabling technologies for detailed DICD implementation, and
three application scenarios of DICD. The road map reveals the common mechanisms
and calculation principles of existing DICD researches, and thus it can provide
systematic guidance for the possible DICD applications that have not been
explored.
| no_new_dataset | 0.935228 |
2302.02615 | Jingyao Li | Jingyao Li, Pengguang Chen, Shaozuo Yu, Zexin He, Shu Liu, Jiaya Jia | Rethinking Out-of-distribution (OOD) Detection: Masked Image Modeling is
All You Need | This paper is accepted by CVPR2023 and our codes are released here:
https://github.com/JulietLJY/MOOD | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The core of out-of-distribution (OOD) detection is to learn the
in-distribution (ID) representation, which is distinguishable from OOD samples.
Previous work applied recognition-based methods to learn the ID features, which
tend to learn shortcuts instead of comprehensive representations. In this work,
we find surprisingly that simply using reconstruction-based methods could boost
the performance of OOD detection significantly. We deeply explore the main
contributors of OOD detection and find that reconstruction-based pretext tasks
have the potential to provide a generally applicable and efficacious prior,
which benefits the model in learning intrinsic data distributions of the ID
dataset. Specifically, we take Masked Image Modeling as a pretext task for our
OOD detection framework (MOOD). Without bells and whistles, MOOD outperforms
previous SOTA of one-class OOD detection by 5.7%, multi-class OOD detection by
3.0%, and near-distribution OOD detection by 2.1%. It even defeats the
10-shot-per-class outlier exposure OOD detection, although we do not include
any OOD samples for our detection
| [
{
"version": "v1",
"created": "Mon, 6 Feb 2023 08:24:41 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 12:46:52 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Li",
"Jingyao",
""
],
[
"Chen",
"Pengguang",
""
],
[
"Yu",
"Shaozuo",
""
],
[
"He",
"Zexin",
""
],
[
"Liu",
"Shu",
""
],
[
"Jia",
"Jiaya",
""
]
] | TITLE: Rethinking Out-of-distribution (OOD) Detection: Masked Image Modeling is
All You Need
ABSTRACT: The core of out-of-distribution (OOD) detection is to learn the
in-distribution (ID) representation, which is distinguishable from OOD samples.
Previous work applied recognition-based methods to learn the ID features, which
tend to learn shortcuts instead of comprehensive representations. In this work,
we find surprisingly that simply using reconstruction-based methods could boost
the performance of OOD detection significantly. We deeply explore the main
contributors of OOD detection and find that reconstruction-based pretext tasks
have the potential to provide a generally applicable and efficacious prior,
which benefits the model in learning intrinsic data distributions of the ID
dataset. Specifically, we take Masked Image Modeling as a pretext task for our
OOD detection framework (MOOD). Without bells and whistles, MOOD outperforms
previous SOTA of one-class OOD detection by 5.7%, multi-class OOD detection by
3.0%, and near-distribution OOD detection by 2.1%. It even defeats the
10-shot-per-class outlier exposure OOD detection, although we do not include
any OOD samples for our detection
| no_new_dataset | 0.943452 |
2302.09498 | Haixu Long | Haixu Long, Xiaolin Zhang, Yanbin Liu, Zongtai Luo, Jianbo Liu | Mutual Exclusive Modulator for Long-Tailed Recognition | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The long-tailed recognition (LTR) is the task of learning high-performance
classifiers given extremely imbalanced training samples between categories.
Most of the existing works address the problem by either enhancing the features
of tail classes or re-balancing the classifiers to reduce the inductive bias.
In this paper, we try to look into the root cause of the LTR task, i.e.,
training samples for each class are greatly imbalanced, and propose a
straightforward solution. We split the categories into three groups, i.e.,
many, medium and few, according to the number of training images. The three
groups of categories are separately predicted to reduce the difficulty for
classification. This idea naturally arises a new problem of how to assign a
given sample to the right class groups? We introduce a mutual exclusive
modulator which can estimate the probability of an image belonging to each
group. Particularly, the modulator consists of a light-weight module and
learned with a mutual exclusive objective. Hence, the output probabilities of
the modulator encode the data volume clues of the training dataset. They are
further utilized as prior information to guide the prediction of the
classifier. We conduct extensive experiments on multiple datasets, e.g.,
ImageNet-LT, Place-LT and iNaturalist 2018 to evaluate the proposed approach.
Our method achieves competitive performance compared to the state-of-the-art
benchmarks.
| [
{
"version": "v1",
"created": "Sun, 19 Feb 2023 07:31:49 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 07:28:14 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Long",
"Haixu",
""
],
[
"Zhang",
"Xiaolin",
""
],
[
"Liu",
"Yanbin",
""
],
[
"Luo",
"Zongtai",
""
],
[
"Liu",
"Jianbo",
""
]
] | TITLE: Mutual Exclusive Modulator for Long-Tailed Recognition
ABSTRACT: The long-tailed recognition (LTR) is the task of learning high-performance
classifiers given extremely imbalanced training samples between categories.
Most of the existing works address the problem by either enhancing the features
of tail classes or re-balancing the classifiers to reduce the inductive bias.
In this paper, we try to look into the root cause of the LTR task, i.e.,
training samples for each class are greatly imbalanced, and propose a
straightforward solution. We split the categories into three groups, i.e.,
many, medium and few, according to the number of training images. The three
groups of categories are separately predicted to reduce the difficulty for
classification. This idea naturally arises a new problem of how to assign a
given sample to the right class groups? We introduce a mutual exclusive
modulator which can estimate the probability of an image belonging to each
group. Particularly, the modulator consists of a light-weight module and
learned with a mutual exclusive objective. Hence, the output probabilities of
the modulator encode the data volume clues of the training dataset. They are
further utilized as prior information to guide the prediction of the
classifier. We conduct extensive experiments on multiple datasets, e.g.,
ImageNet-LT, Place-LT and iNaturalist 2018 to evaluate the proposed approach.
Our method achieves competitive performance compared to the state-of-the-art
benchmarks.
| no_new_dataset | 0.947381 |
2302.10834 | Sanat Ramesh | Sanat Ramesh, Diego Dall'Alba, Cristians Gonzalez, Tong Yu, Pietro
Mascagni, Didier Mutter, Jacques Marescaux, Paolo Fiorini, and Nicolas Padoy | Weakly Supervised Temporal Convolutional Networks for Fine-grained
Surgical Activity Recognition | null | null | 10.1109/TMI.2023.3262847 | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Automatic recognition of fine-grained surgical activities, called steps, is a
challenging but crucial task for intelligent intra-operative computer
assistance. The development of current vision-based activity recognition
methods relies heavily on a high volume of manually annotated data. This data
is difficult and time-consuming to generate and requires domain-specific
knowledge. In this work, we propose to use coarser and easier-to-annotate
activity labels, namely phases, as weak supervision to learn step recognition
with fewer step annotated videos. We introduce a step-phase dependency loss to
exploit the weak supervision signal. We then employ a Single-Stage Temporal
Convolutional Network (SS-TCN) with a ResNet-50 backbone, trained in an
end-to-end fashion from weakly annotated videos, for temporal activity
segmentation and recognition. We extensively evaluate and show the
effectiveness of the proposed method on a large video dataset consisting of 40
laparoscopic gastric bypass procedures and the public benchmark CATARACTS
containing 50 cataract surgeries.
| [
{
"version": "v1",
"created": "Tue, 21 Feb 2023 17:26:49 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 13:55:28 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Ramesh",
"Sanat",
""
],
[
"Dall'Alba",
"Diego",
""
],
[
"Gonzalez",
"Cristians",
""
],
[
"Yu",
"Tong",
""
],
[
"Mascagni",
"Pietro",
""
],
[
"Mutter",
"Didier",
""
],
[
"Marescaux",
"Jacques",
""
],
[
"Fiorini",
"Paolo",
""
],
[
"Padoy",
"Nicolas",
""
]
] | TITLE: Weakly Supervised Temporal Convolutional Networks for Fine-grained
Surgical Activity Recognition
ABSTRACT: Automatic recognition of fine-grained surgical activities, called steps, is a
challenging but crucial task for intelligent intra-operative computer
assistance. The development of current vision-based activity recognition
methods relies heavily on a high volume of manually annotated data. This data
is difficult and time-consuming to generate and requires domain-specific
knowledge. In this work, we propose to use coarser and easier-to-annotate
activity labels, namely phases, as weak supervision to learn step recognition
with fewer step annotated videos. We introduce a step-phase dependency loss to
exploit the weak supervision signal. We then employ a Single-Stage Temporal
Convolutional Network (SS-TCN) with a ResNet-50 backbone, trained in an
end-to-end fashion from weakly annotated videos, for temporal activity
segmentation and recognition. We extensively evaluate and show the
effectiveness of the proposed method on a large video dataset consisting of 40
laparoscopic gastric bypass procedures and the public benchmark CATARACTS
containing 50 cataract surgeries.
| new_dataset | 0.538028 |
2302.12604 | Samuel Holt | Samuel Holt, Alihan H\"uy\"uk, Zhaozhi Qian, Hao Sun, Mihaela van der
Schaar | Neural Laplace Control for Continuous-time Delayed Systems | Proceedings of the 26th International Conference on Artificial
Intelligence and Statistics (AISTATS) 2023, Valencia, Spain. PMLR: Volume
206. Copyright 2023 by the author(s) | null | null | null | cs.LG cs.AI cs.RO stat.ML | http://creativecommons.org/licenses/by/4.0/ | Many real-world offline reinforcement learning (RL) problems involve
continuous-time environments with delays. Such environments are characterized
by two distinctive features: firstly, the state x(t) is observed at irregular
time intervals, and secondly, the current action a(t) only affects the future
state x(t + g) with an unknown delay g > 0. A prime example of such an
environment is satellite control where the communication link between earth and
a satellite causes irregular observations and delays. Existing offline RL
algorithms have achieved success in environments with irregularly observed
states in time or known delays. However, environments involving both irregular
observations in time and unknown delays remains an open and challenging
problem. To this end, we propose Neural Laplace Control, a continuous-time
model-based offline RL method that combines a Neural Laplace dynamics model
with a model predictive control (MPC) planner--and is able to learn from an
offline dataset sampled with irregular time intervals from an environment that
has a inherent unknown constant delay. We show experimentally on
continuous-time delayed environments it is able to achieve near expert policy
performance.
| [
{
"version": "v1",
"created": "Fri, 24 Feb 2023 12:40:28 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 01:28:57 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Holt",
"Samuel",
""
],
[
"Hüyük",
"Alihan",
""
],
[
"Qian",
"Zhaozhi",
""
],
[
"Sun",
"Hao",
""
],
[
"van der Schaar",
"Mihaela",
""
]
] | TITLE: Neural Laplace Control for Continuous-time Delayed Systems
ABSTRACT: Many real-world offline reinforcement learning (RL) problems involve
continuous-time environments with delays. Such environments are characterized
by two distinctive features: firstly, the state x(t) is observed at irregular
time intervals, and secondly, the current action a(t) only affects the future
state x(t + g) with an unknown delay g > 0. A prime example of such an
environment is satellite control where the communication link between earth and
a satellite causes irregular observations and delays. Existing offline RL
algorithms have achieved success in environments with irregularly observed
states in time or known delays. However, environments involving both irregular
observations in time and unknown delays remains an open and challenging
problem. To this end, we propose Neural Laplace Control, a continuous-time
model-based offline RL method that combines a Neural Laplace dynamics model
with a model predictive control (MPC) planner--and is able to learn from an
offline dataset sampled with irregular time intervals from an environment that
has a inherent unknown constant delay. We show experimentally on
continuous-time delayed environments it is able to achieve near expert policy
performance.
| no_new_dataset | 0.951549 |
2303.02660 | Meiling Fang | Meiling Fang and Marco Huber and Naser Damer | SynthASpoof: Developing Face Presentation Attack Detection Based on
Privacy-friendly Synthetic Data | Accepted at CVPR workshop 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Recently, significant progress has been made in face presentation attack
detection (PAD), which aims to secure face recognition systems against
presentation attacks, owing to the availability of several face PAD datasets.
However, all available datasets are based on privacy and legally-sensitive
authentic biometric data with a limited number of subjects. To target these
legal and technical challenges, this work presents the first synthetic-based
face PAD dataset, named SynthASpoof, as a large-scale PAD development dataset.
The bona fide samples in SynthASpoof are synthetically generated and the attack
samples are collected by presenting such synthetic data to capture systems in a
real attack scenario. The experimental results demonstrate the feasibility of
using SynthASpoof for the development of face PAD. Moreover, we boost the
performance of such a solution by incorporating the domain generalization tool
MixStyle into the PAD solutions. Additionally, we showed the viability of using
synthetic data as a supplement to enrich the diversity of limited authentic
training data and consistently enhance PAD performances. The SynthASpoof
dataset, containing 25,000 bona fide and 78,800 attack samples, the
implementation, and the pre-trained weights are made publicly available.
| [
{
"version": "v1",
"created": "Sun, 5 Mar 2023 12:35:58 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 09:38:16 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Fang",
"Meiling",
""
],
[
"Huber",
"Marco",
""
],
[
"Damer",
"Naser",
""
]
] | TITLE: SynthASpoof: Developing Face Presentation Attack Detection Based on
Privacy-friendly Synthetic Data
ABSTRACT: Recently, significant progress has been made in face presentation attack
detection (PAD), which aims to secure face recognition systems against
presentation attacks, owing to the availability of several face PAD datasets.
However, all available datasets are based on privacy and legally-sensitive
authentic biometric data with a limited number of subjects. To target these
legal and technical challenges, this work presents the first synthetic-based
face PAD dataset, named SynthASpoof, as a large-scale PAD development dataset.
The bona fide samples in SynthASpoof are synthetically generated and the attack
samples are collected by presenting such synthetic data to capture systems in a
real attack scenario. The experimental results demonstrate the feasibility of
using SynthASpoof for the development of face PAD. Moreover, we boost the
performance of such a solution by incorporating the domain generalization tool
MixStyle into the PAD solutions. Additionally, we showed the viability of using
synthetic data as a supplement to enrich the diversity of limited authentic
training data and consistently enhance PAD performances. The SynthASpoof
dataset, containing 25,000 bona fide and 78,800 attack samples, the
implementation, and the pre-trained weights are made publicly available.
| new_dataset | 0.964623 |
2303.04681 | Sungho Shin | Sungho Shin, Yeonguk Yu, Kyoobin Lee | Enhancing Low-resolution Face Recognition with Feature Similarity
Knowledge Distillation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | In this study, we introduce a feature knowledge distillation framework to
improve low-resolution (LR) face recognition performance using knowledge
obtained from high-resolution (HR) images. The proposed framework transfers
informative features from an HR-trained network to an LR-trained network by
reducing the distance between them. A cosine similarity measure was employed as
a distance metric to effectively align the HR and LR features. This approach
differs from conventional knowledge distillation frameworks, which use the L_p
distance metrics and offer the advantage of converging well when reducing the
distance between features of different resolutions. Our framework achieved a 3%
improvement over the previous state-of-the-art method on the AgeDB-30 benchmark
without bells and whistles, while maintaining a strong performance on HR
images. The effectiveness of cosine similarity as a distance metric was
validated through statistical analysis, making our approach a promising
solution for real-world applications in which LR images are frequently
encountered. The code and pretrained models are publicly available on
https://github.com/gist-ailab/feature-similarity-KD.
| [
{
"version": "v1",
"created": "Wed, 8 Mar 2023 16:11:46 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 00:31:44 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Shin",
"Sungho",
""
],
[
"Yu",
"Yeonguk",
""
],
[
"Lee",
"Kyoobin",
""
]
] | TITLE: Enhancing Low-resolution Face Recognition with Feature Similarity
Knowledge Distillation
ABSTRACT: In this study, we introduce a feature knowledge distillation framework to
improve low-resolution (LR) face recognition performance using knowledge
obtained from high-resolution (HR) images. The proposed framework transfers
informative features from an HR-trained network to an LR-trained network by
reducing the distance between them. A cosine similarity measure was employed as
a distance metric to effectively align the HR and LR features. This approach
differs from conventional knowledge distillation frameworks, which use the L_p
distance metrics and offer the advantage of converging well when reducing the
distance between features of different resolutions. Our framework achieved a 3%
improvement over the previous state-of-the-art method on the AgeDB-30 benchmark
without bells and whistles, while maintaining a strong performance on HR
images. The effectiveness of cosine similarity as a distance metric was
validated through statistical analysis, making our approach a promising
solution for real-world applications in which LR images are frequently
encountered. The code and pretrained models are publicly available on
https://github.com/gist-ailab/feature-similarity-KD.
| no_new_dataset | 0.951414 |
2303.08498 | Lei Yang | Lei Yang, Kaicheng Yu, Tao Tang, Jun Li, Kun Yuan, Li Wang, Xinyu
Zhang, Peng Chen | BEVHeight: A Robust Framework for Vision-based Roadside 3D Object
Detection | Accepted by CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While most recent autonomous driving system focuses on developing perception
methods on ego-vehicle sensors, people tend to overlook an alternative approach
to leverage intelligent roadside cameras to extend the perception ability
beyond the visual range. We discover that the state-of-the-art vision-centric
bird's eye view detection methods have inferior performances on roadside
cameras. This is because these methods mainly focus on recovering the depth
regarding the camera center, where the depth difference between the car and the
ground quickly shrinks while the distance increases. In this paper, we propose
a simple yet effective approach, dubbed BEVHeight, to address this issue. In
essence, instead of predicting the pixel-wise depth, we regress the height to
the ground to achieve a distance-agnostic formulation to ease the optimization
process of camera-only perception methods. On popular 3D detection benchmarks
of roadside cameras, our method surpasses all previous vision-centric methods
by a significant margin. The code is available at
{\url{https://github.com/ADLab-AutoDrive/BEVHeight}}.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 10:18:53 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 07:19:13 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Yang",
"Lei",
""
],
[
"Yu",
"Kaicheng",
""
],
[
"Tang",
"Tao",
""
],
[
"Li",
"Jun",
""
],
[
"Yuan",
"Kun",
""
],
[
"Wang",
"Li",
""
],
[
"Zhang",
"Xinyu",
""
],
[
"Chen",
"Peng",
""
]
] | TITLE: BEVHeight: A Robust Framework for Vision-based Roadside 3D Object
Detection
ABSTRACT: While most recent autonomous driving system focuses on developing perception
methods on ego-vehicle sensors, people tend to overlook an alternative approach
to leverage intelligent roadside cameras to extend the perception ability
beyond the visual range. We discover that the state-of-the-art vision-centric
bird's eye view detection methods have inferior performances on roadside
cameras. This is because these methods mainly focus on recovering the depth
regarding the camera center, where the depth difference between the car and the
ground quickly shrinks while the distance increases. In this paper, we propose
a simple yet effective approach, dubbed BEVHeight, to address this issue. In
essence, instead of predicting the pixel-wise depth, we regress the height to
the ground to achieve a distance-agnostic formulation to ease the optimization
process of camera-only perception methods. On popular 3D detection benchmarks
of roadside cameras, our method surpasses all previous vision-centric methods
by a significant margin. The code is available at
{\url{https://github.com/ADLab-AutoDrive/BEVHeight}}.
| no_new_dataset | 0.946695 |
2303.10849 | Wei Zhang | Wei Zhang, Bowen Ma, Feng Qiu, Yu Ding | Multi-modal Facial Affective Analysis based on Masked Autoencoder | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Human affective behavior analysis focuses on analyzing human expressions or
other behaviors to enhance the understanding of human psychology. The CVPR 2023
Competition on Affective Behavior Analysis in-the-wild (ABAW) is dedicated to
providing high-quality and large-scale Aff-wild2 for the recognition of
commonly used emotion representations, such as Action Units (AU), basic
expression categories(EXPR), and Valence-Arousal (VA). The competition is
committed to making significant strides in improving the accuracy and
practicality of affective analysis research in real-world scenarios. In this
paper, we introduce our submission to the CVPR 2023: ABAW5. Our approach
involves several key components. First, we utilize the visual information from
a Masked Autoencoder(MAE) model that has been pre-trained on a large-scale face
image dataset in a self-supervised manner. Next, we finetune the MAE encoder on
the image frames from the Aff-wild2 for AU, EXPR and VA tasks, which can be
regarded as a static and uni-modal training. Additionally, we leverage the
multi-modal and temporal information from the videos and implement a
transformer-based framework to fuse the multi-modal features. Our approach
achieves impressive results in the ABAW5 competition, with an average F1 score
of 55.49\% and 41.21\% in the AU and EXPR tracks, respectively, and an average
CCC of 0.6372 in the VA track. Our approach ranks first in the EXPR and AU
tracks, and second in the VA track. Extensive quantitative experiments and
ablation studies demonstrate the effectiveness of our proposed method.
| [
{
"version": "v1",
"created": "Mon, 20 Mar 2023 03:58:03 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 06:41:45 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Zhang",
"Wei",
""
],
[
"Ma",
"Bowen",
""
],
[
"Qiu",
"Feng",
""
],
[
"Ding",
"Yu",
""
]
] | TITLE: Multi-modal Facial Affective Analysis based on Masked Autoencoder
ABSTRACT: Human affective behavior analysis focuses on analyzing human expressions or
other behaviors to enhance the understanding of human psychology. The CVPR 2023
Competition on Affective Behavior Analysis in-the-wild (ABAW) is dedicated to
providing high-quality and large-scale Aff-wild2 for the recognition of
commonly used emotion representations, such as Action Units (AU), basic
expression categories(EXPR), and Valence-Arousal (VA). The competition is
committed to making significant strides in improving the accuracy and
practicality of affective analysis research in real-world scenarios. In this
paper, we introduce our submission to the CVPR 2023: ABAW5. Our approach
involves several key components. First, we utilize the visual information from
a Masked Autoencoder(MAE) model that has been pre-trained on a large-scale face
image dataset in a self-supervised manner. Next, we finetune the MAE encoder on
the image frames from the Aff-wild2 for AU, EXPR and VA tasks, which can be
regarded as a static and uni-modal training. Additionally, we leverage the
multi-modal and temporal information from the videos and implement a
transformer-based framework to fuse the multi-modal features. Our approach
achieves impressive results in the ABAW5 competition, with an average F1 score
of 55.49\% and 41.21\% in the AU and EXPR tracks, respectively, and an average
CCC of 0.6372 in the VA track. Our approach ranks first in the EXPR and AU
tracks, and second in the VA track. Extensive quantitative experiments and
ablation studies demonstrate the effectiveness of our proposed method.
| no_new_dataset | 0.951459 |
2303.13065 | Xinnian Liang | Xinnian Liang, Shuangzhi Wu, Hui Huang, Jiaqi Bai, Chao Bian, Zhoujun
Li | Retrieval-Augmented Classification with Decoupled Representation | preprint | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Retrieval augmented methods have shown promising results in various
classification tasks. However, existing methods focus on retrieving extra
context to enrich the input, which is noise sensitive and non-expandable. In
this paper, following this line, we propose a $k$-nearest-neighbor (KNN) -based
method for retrieval augmented classifications, which interpolates the
predicted label distribution with retrieved instances' label distributions.
Different from the standard KNN process, we propose a decoupling mechanism as
we find that shared representation for classification and retrieval hurts
performance and leads to training instability. We evaluate our method on a wide
range of classification datasets. Experimental results demonstrate the
effectiveness and robustness of our proposed method. We also conduct extra
experiments to analyze the contributions of different components in our
model.\footnote{\url{https://github.com/xnliang98/knn-cls-w-decoupling}}
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 06:33:06 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 09:13:30 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Liang",
"Xinnian",
""
],
[
"Wu",
"Shuangzhi",
""
],
[
"Huang",
"Hui",
""
],
[
"Bai",
"Jiaqi",
""
],
[
"Bian",
"Chao",
""
],
[
"Li",
"Zhoujun",
""
]
] | TITLE: Retrieval-Augmented Classification with Decoupled Representation
ABSTRACT: Retrieval augmented methods have shown promising results in various
classification tasks. However, existing methods focus on retrieving extra
context to enrich the input, which is noise sensitive and non-expandable. In
this paper, following this line, we propose a $k$-nearest-neighbor (KNN) -based
method for retrieval augmented classifications, which interpolates the
predicted label distribution with retrieved instances' label distributions.
Different from the standard KNN process, we propose a decoupling mechanism as
we find that shared representation for classification and retrieval hurts
performance and leads to training instability. We evaluate our method on a wide
range of classification datasets. Experimental results demonstrate the
effectiveness and robustness of our proposed method. We also conduct extra
experiments to analyze the contributions of different components in our
model.\footnote{\url{https://github.com/xnliang98/knn-cls-w-decoupling}}
| no_new_dataset | 0.947866 |
2303.16870 | Daniel Cajueiro | Nuno Crokidakis, Marcio Argollo de Menezes and Daniel O. Cajueiro | Questions of science: chatting with ChatGPT about complex systems | This is a work in progress | null | null | null | physics.soc-ph cs.AI cs.HC | http://creativecommons.org/licenses/by/4.0/ | We present an overview of the complex systems field using ChatGPT as a
representation of the community's understanding. ChatGPT has learned language
patterns and styles from a large dataset of internet texts, allowing it to
provide answers that reflect common opinions, ideas, and language patterns
found in the community. Our exploration covers both teaching and learning, and
research topics. We recognize the value of ChatGPT as a source for the
community's ideas.
| [
{
"version": "v1",
"created": "Wed, 29 Mar 2023 17:27:05 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Crokidakis",
"Nuno",
""
],
[
"de Menezes",
"Marcio Argollo",
""
],
[
"Cajueiro",
"Daniel O.",
""
]
] | TITLE: Questions of science: chatting with ChatGPT about complex systems
ABSTRACT: We present an overview of the complex systems field using ChatGPT as a
representation of the community's understanding. ChatGPT has learned language
patterns and styles from a large dataset of internet texts, allowing it to
provide answers that reflect common opinions, ideas, and language patterns
found in the community. Our exploration covers both teaching and learning, and
research topics. We recognize the value of ChatGPT as a source for the
community's ideas.
| no_new_dataset | 0.933854 |
2304.04103 | Cristian Meo | Cristian Meo, Anirudh Goyal and Justin Dauwels | TC-VAE: Uncovering Out-of-Distribution Data Generative Factors | The paper is incomplete as it is. We will work on it and repost it | null | null | null | cs.LG cs.AI cs.IT math.IT | http://creativecommons.org/licenses/by/4.0/ | Uncovering data generative factors is the ultimate goal of disentanglement
learning. Although many works proposed disentangling generative models able to
uncover the underlying generative factors of a dataset, so far no one was able
to uncover OOD generative factors (i.e., factors of variations that are not
explicitly shown on the dataset). Moreover, the datasets used to validate these
models are synthetically generated using a balanced mixture of some predefined
generative factors, implicitly assuming that generative factors are uniformly
distributed across the datasets. However, real datasets do not present this
property. In this work we analyse the effect of using datasets with unbalanced
generative factors, providing qualitative and quantitative results for widely
used generative models. Moreover, we propose TC-VAE, a generative model
optimized using a lower bound of the joint total correlation between the
learned latent representations and the input data. We show that the proposed
model is able to uncover OOD generative factors on different datasets and
outperforms on average the related baselines in terms of downstream
disentanglement metrics.
| [
{
"version": "v1",
"created": "Sat, 8 Apr 2023 21:16:46 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 00:45:46 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Meo",
"Cristian",
""
],
[
"Goyal",
"Anirudh",
""
],
[
"Dauwels",
"Justin",
""
]
] | TITLE: TC-VAE: Uncovering Out-of-Distribution Data Generative Factors
ABSTRACT: Uncovering data generative factors is the ultimate goal of disentanglement
learning. Although many works proposed disentangling generative models able to
uncover the underlying generative factors of a dataset, so far no one was able
to uncover OOD generative factors (i.e., factors of variations that are not
explicitly shown on the dataset). Moreover, the datasets used to validate these
models are synthetically generated using a balanced mixture of some predefined
generative factors, implicitly assuming that generative factors are uniformly
distributed across the datasets. However, real datasets do not present this
property. In this work we analyse the effect of using datasets with unbalanced
generative factors, providing qualitative and quantitative results for widely
used generative models. Moreover, we propose TC-VAE, a generative model
optimized using a lower bound of the joint total correlation between the
learned latent representations and the input data. We show that the proposed
model is able to uncover OOD generative factors on different datasets and
outperforms on average the related baselines in terms of downstream
disentanglement metrics.
| no_new_dataset | 0.944177 |
2304.04385 | Alexander Toshev | Brandon McKinzie, Joseph Cheng, Vaishaal Shankar, Yinfei Yang,
Jonathon Shlens, Alexander Toshev | On Robustness in Multimodal Learning | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Multimodal learning is defined as learning over multiple heterogeneous input
modalities such as video, audio, and text. In this work, we are concerned with
understanding how models behave as the type of modalities differ between
training and deployment, a situation that naturally arises in many applications
of multimodal learning to hardware platforms. We present a multimodal
robustness framework to provide a systematic analysis of common multimodal
representation learning methods. Further, we identify robustness short-comings
of these approaches and propose two intervention techniques leading to
$1.5\times$-$4\times$ robustness improvements on three datasets, AudioSet,
Kinetics-400 and ImageNet-Captions. Finally, we demonstrate that these
interventions better utilize additional modalities, if present, to achieve
competitive results of $44.2$ mAP on AudioSet 20K.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 05:02:07 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 01:23:01 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"McKinzie",
"Brandon",
""
],
[
"Cheng",
"Joseph",
""
],
[
"Shankar",
"Vaishaal",
""
],
[
"Yang",
"Yinfei",
""
],
[
"Shlens",
"Jonathon",
""
],
[
"Toshev",
"Alexander",
""
]
] | TITLE: On Robustness in Multimodal Learning
ABSTRACT: Multimodal learning is defined as learning over multiple heterogeneous input
modalities such as video, audio, and text. In this work, we are concerned with
understanding how models behave as the type of modalities differ between
training and deployment, a situation that naturally arises in many applications
of multimodal learning to hardware platforms. We present a multimodal
robustness framework to provide a systematic analysis of common multimodal
representation learning methods. Further, we identify robustness short-comings
of these approaches and propose two intervention techniques leading to
$1.5\times$-$4\times$ robustness improvements on three datasets, AudioSet,
Kinetics-400 and ImageNet-Captions. Finally, we demonstrate that these
interventions better utilize additional modalities, if present, to achieve
competitive results of $44.2$ mAP on AudioSet 20K.
| no_new_dataset | 0.945197 |
2304.04598 | Lequn Chen | Lequn Chen, Xiling Yao, Chaolin Tan, Weiyang He, Jinlong Su, Fei Weng,
Youxiang Chew, Nicholas Poh Huat Ng, Seung Ki Moon | In-situ crack and keyhole pore detection in laser directed energy
deposition through acoustic signal and deep learning | 36 Pages, 16 Figures, accepted at journal Additive Manufacturing | null | 10.1016/j.addma.2023.103547 | null | cs.SD eess.AS eess.SP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Cracks and keyhole pores are detrimental defects in alloys produced by laser
directed energy deposition (LDED). Laser-material interaction sound may hold
information about underlying complex physical events such as crack propagation
and pores formation. However, due to the noisy environment and intricate signal
content, acoustic-based monitoring in LDED has received little attention. This
paper proposes a novel acoustic-based in-situ defect detection strategy in
LDED. The key contribution of this study is to develop an in-situ acoustic
signal denoising, feature extraction, and sound classification pipeline that
incorporates convolutional neural networks (CNN) for online defect prediction.
Microscope images are used to identify locations of the cracks and keyhole
pores within a part. The defect locations are spatiotemporally registered with
acoustic signal. Various acoustic features corresponding to defect-free
regions, cracks, and keyhole pores are extracted and analysed in time-domain,
frequency-domain, and time-frequency representations. The CNN model is trained
to predict defect occurrences using the Mel-Frequency Cepstral Coefficients
(MFCCs) of the lasermaterial interaction sound. The CNN model is compared to
various classic machine learning models trained on the denoised acoustic
dataset and raw acoustic dataset. The validation results shows that the CNN
model trained on the denoised dataset outperforms others with the highest
overall accuracy (89%), keyhole pore prediction accuracy (93%), and AUC-ROC
score (98%). Furthermore, the trained CNN model can be deployed into an
in-house developed software platform for online quality monitoring. The
proposed strategy is the first study to use acoustic signals with deep learning
for insitu defect detection in LDED process.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 14:12:11 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Chen",
"Lequn",
""
],
[
"Yao",
"Xiling",
""
],
[
"Tan",
"Chaolin",
""
],
[
"He",
"Weiyang",
""
],
[
"Su",
"Jinlong",
""
],
[
"Weng",
"Fei",
""
],
[
"Chew",
"Youxiang",
""
],
[
"Ng",
"Nicholas Poh Huat",
""
],
[
"Moon",
"Seung Ki",
""
]
] | TITLE: In-situ crack and keyhole pore detection in laser directed energy
deposition through acoustic signal and deep learning
ABSTRACT: Cracks and keyhole pores are detrimental defects in alloys produced by laser
directed energy deposition (LDED). Laser-material interaction sound may hold
information about underlying complex physical events such as crack propagation
and pores formation. However, due to the noisy environment and intricate signal
content, acoustic-based monitoring in LDED has received little attention. This
paper proposes a novel acoustic-based in-situ defect detection strategy in
LDED. The key contribution of this study is to develop an in-situ acoustic
signal denoising, feature extraction, and sound classification pipeline that
incorporates convolutional neural networks (CNN) for online defect prediction.
Microscope images are used to identify locations of the cracks and keyhole
pores within a part. The defect locations are spatiotemporally registered with
acoustic signal. Various acoustic features corresponding to defect-free
regions, cracks, and keyhole pores are extracted and analysed in time-domain,
frequency-domain, and time-frequency representations. The CNN model is trained
to predict defect occurrences using the Mel-Frequency Cepstral Coefficients
(MFCCs) of the lasermaterial interaction sound. The CNN model is compared to
various classic machine learning models trained on the denoised acoustic
dataset and raw acoustic dataset. The validation results shows that the CNN
model trained on the denoised dataset outperforms others with the highest
overall accuracy (89%), keyhole pore prediction accuracy (93%), and AUC-ROC
score (98%). Furthermore, the trained CNN model can be deployed into an
in-house developed software platform for online quality monitoring. The
proposed strategy is the first study to use acoustic signals with deep learning
for insitu defect detection in LDED process.
| no_new_dataset | 0.957517 |
2304.04709 | Lv Tang | Lv Tang, Haoke Xiao, Bo Li | Can SAM Segment Anything? When SAM Meets Camouflaged Object Detection | null | null | null | null | cs.CV | http://creativecommons.org/publicdomain/zero/1.0/ | SAM is a segmentation model recently released by Meta AI Research and has
been gaining attention quickly due to its impressive performance in generic
object segmentation. However, its ability to generalize to specific scenes such
as camouflaged scenes is still unknown. Camouflaged object detection (COD)
involves identifying objects that are seamlessly integrated into their
surroundings and has numerous practical applications in fields such as
medicine, art, and agriculture. In this study, we try to ask if SAM can address
the COD task and evaluate the performance of SAM on the COD benchmark by
employing maximum segmentation evaluation and camouflage location evaluation.
We also compare SAM's performance with 22 state-of-the-art COD methods. Our
results indicate that while SAM shows promise in generic object segmentation,
its performance on the COD task is limited. This presents an opportunity for
further research to explore how to build a stronger SAM that may address the
COD task. The results of this paper are provided in
\url{https://github.com/luckybird1994/SAMCOD}.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 17:05:58 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Apr 2023 03:53:13 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Tang",
"Lv",
""
],
[
"Xiao",
"Haoke",
""
],
[
"Li",
"Bo",
""
]
] | TITLE: Can SAM Segment Anything? When SAM Meets Camouflaged Object Detection
ABSTRACT: SAM is a segmentation model recently released by Meta AI Research and has
been gaining attention quickly due to its impressive performance in generic
object segmentation. However, its ability to generalize to specific scenes such
as camouflaged scenes is still unknown. Camouflaged object detection (COD)
involves identifying objects that are seamlessly integrated into their
surroundings and has numerous practical applications in fields such as
medicine, art, and agriculture. In this study, we try to ask if SAM can address
the COD task and evaluate the performance of SAM on the COD benchmark by
employing maximum segmentation evaluation and camouflage location evaluation.
We also compare SAM's performance with 22 state-of-the-art COD methods. Our
results indicate that while SAM shows promise in generic object segmentation,
its performance on the COD task is limited. This presents an opportunity for
further research to explore how to build a stronger SAM that may address the
COD task. The results of this paper are provided in
\url{https://github.com/luckybird1994/SAMCOD}.
| no_new_dataset | 0.947914 |
2304.04751 | Opeoluwa Owoyele | Eloghosa Ikponmwoba and Ope Owoyele | DeepHive: A multi-agent reinforcement learning approach for automated
discovery of swarm-based optimization policies | null | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present an approach for designing swarm-based optimizers for the global
optimization of expensive black-box functions. In the proposed approach, the
problem of finding efficient optimizers is framed as a reinforcement learning
problem, where the goal is to find optimization policies that require a few
function evaluations to converge to the global optimum. The state of each agent
within the swarm is defined as its current position and function value within a
design space and the agents learn to take favorable actions that maximize
reward, which is based on the final value of the objective function. The
proposed approach is tested on various benchmark optimization functions and
compared to the performance of other global optimization strategies.
Furthermore, the effect of changing the number of agents, as well as the
generalization capabilities of the trained agents are investigated. The results
show superior performance compared to the other optimizers, desired scaling
when the number of agents is varied, and acceptable performance even when
applied to unseen functions. On a broader scale, the results show promise for
the rapid development of domain-specific optimizers.
| [
{
"version": "v1",
"created": "Wed, 29 Mar 2023 18:08:08 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Ikponmwoba",
"Eloghosa",
""
],
[
"Owoyele",
"Ope",
""
]
] | TITLE: DeepHive: A multi-agent reinforcement learning approach for automated
discovery of swarm-based optimization policies
ABSTRACT: We present an approach for designing swarm-based optimizers for the global
optimization of expensive black-box functions. In the proposed approach, the
problem of finding efficient optimizers is framed as a reinforcement learning
problem, where the goal is to find optimization policies that require a few
function evaluations to converge to the global optimum. The state of each agent
within the swarm is defined as its current position and function value within a
design space and the agents learn to take favorable actions that maximize
reward, which is based on the final value of the objective function. The
proposed approach is tested on various benchmark optimization functions and
compared to the performance of other global optimization strategies.
Furthermore, the effect of changing the number of agents, as well as the
generalization capabilities of the trained agents are investigated. The results
show superior performance compared to the other optimizers, desired scaling
when the number of agents is varied, and acceptable performance even when
applied to unseen functions. On a broader scale, the results show promise for
the rapid development of domain-specific optimizers.
| no_new_dataset | 0.947088 |
2304.04774 | Xiao Wu | ZiHan Cao, ShiQi Cao, Xiao Wu, JunMing Hou, Ran Ran, Liang-Jian Deng | DDRF: Denoising Diffusion Model for Remote Sensing Image Fusion | null | null | null | null | cs.CV cs.AI eess.IV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Denosing diffusion model, as a generative model, has received a lot of
attention in the field of image generation recently, thanks to its powerful
generation capability. However, diffusion models have not yet received
sufficient research in the field of image fusion. In this article, we introduce
diffusion model to the image fusion field, treating the image fusion task as
image-to-image translation and designing two different conditional injection
modulation modules (i.e., style transfer modulation and wavelet modulation) to
inject coarse-grained style information and fine-grained high-frequency and
low-frequency information into the diffusion UNet, thereby generating fused
images. In addition, we also discussed the residual learning and the selection
of training objectives of the diffusion model in the image fusion task.
Extensive experimental results based on quantitative and qualitative
assessments compared with benchmarks demonstrates state-of-the-art results and
good generalization performance in image fusion tasks. Finally, it is hoped
that our method can inspire other works and gain insight into this field to
better apply the diffusion model to image fusion tasks. Code shall be released
for better reproducibility.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 12:28:27 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Cao",
"ZiHan",
""
],
[
"Cao",
"ShiQi",
""
],
[
"Wu",
"Xiao",
""
],
[
"Hou",
"JunMing",
""
],
[
"Ran",
"Ran",
""
],
[
"Deng",
"Liang-Jian",
""
]
] | TITLE: DDRF: Denoising Diffusion Model for Remote Sensing Image Fusion
ABSTRACT: Denosing diffusion model, as a generative model, has received a lot of
attention in the field of image generation recently, thanks to its powerful
generation capability. However, diffusion models have not yet received
sufficient research in the field of image fusion. In this article, we introduce
diffusion model to the image fusion field, treating the image fusion task as
image-to-image translation and designing two different conditional injection
modulation modules (i.e., style transfer modulation and wavelet modulation) to
inject coarse-grained style information and fine-grained high-frequency and
low-frequency information into the diffusion UNet, thereby generating fused
images. In addition, we also discussed the residual learning and the selection
of training objectives of the diffusion model in the image fusion task.
Extensive experimental results based on quantitative and qualitative
assessments compared with benchmarks demonstrates state-of-the-art results and
good generalization performance in image fusion tasks. Finally, it is hoped
that our method can inspire other works and gain insight into this field to
better apply the diffusion model to image fusion tasks. Code shall be released
for better reproducibility.
| no_new_dataset | 0.950273 |
2304.04779 | Zhenyu Hou | Zhenyu Hou, Yufei He, Yukuo Cen, Xiao Liu, Yuxiao Dong, Evgeny
Kharlamov, Jie Tang | GraphMAE2: A Decoding-Enhanced Masked Self-Supervised Graph Learner | Accepted to WWW'23 | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Graph self-supervised learning (SSL), including contrastive and generative
approaches, offers great potential to address the fundamental challenge of
label scarcity in real-world graph data. Among both sets of graph SSL
techniques, the masked graph autoencoders (e.g., GraphMAE)--one type of
generative method--have recently produced promising results. The idea behind
this is to reconstruct the node features (or structures)--that are randomly
masked from the input--with the autoencoder architecture. However, the
performance of masked feature reconstruction naturally relies on the
discriminability of the input features and is usually vulnerable to disturbance
in the features. In this paper, we present a masked self-supervised learning
framework GraphMAE2 with the goal of overcoming this issue. The idea is to
impose regularization on feature reconstruction for graph SSL. Specifically, we
design the strategies of multi-view random re-mask decoding and latent
representation prediction to regularize the feature reconstruction. The
multi-view random re-mask decoding is to introduce randomness into
reconstruction in the feature space, while the latent representation prediction
is to enforce the reconstruction in the embedding space. Extensive experiments
show that GraphMAE2 can consistently generate top results on various public
datasets, including at least 2.45% improvements over state-of-the-art baselines
on ogbn-Papers100M with 111M nodes and 1.6B edges.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 17:25:50 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Hou",
"Zhenyu",
""
],
[
"He",
"Yufei",
""
],
[
"Cen",
"Yukuo",
""
],
[
"Liu",
"Xiao",
""
],
[
"Dong",
"Yuxiao",
""
],
[
"Kharlamov",
"Evgeny",
""
],
[
"Tang",
"Jie",
""
]
] | TITLE: GraphMAE2: A Decoding-Enhanced Masked Self-Supervised Graph Learner
ABSTRACT: Graph self-supervised learning (SSL), including contrastive and generative
approaches, offers great potential to address the fundamental challenge of
label scarcity in real-world graph data. Among both sets of graph SSL
techniques, the masked graph autoencoders (e.g., GraphMAE)--one type of
generative method--have recently produced promising results. The idea behind
this is to reconstruct the node features (or structures)--that are randomly
masked from the input--with the autoencoder architecture. However, the
performance of masked feature reconstruction naturally relies on the
discriminability of the input features and is usually vulnerable to disturbance
in the features. In this paper, we present a masked self-supervised learning
framework GraphMAE2 with the goal of overcoming this issue. The idea is to
impose regularization on feature reconstruction for graph SSL. Specifically, we
design the strategies of multi-view random re-mask decoding and latent
representation prediction to regularize the feature reconstruction. The
multi-view random re-mask decoding is to introduce randomness into
reconstruction in the feature space, while the latent representation prediction
is to enforce the reconstruction in the embedding space. Extensive experiments
show that GraphMAE2 can consistently generate top results on various public
datasets, including at least 2.45% improvements over state-of-the-art baselines
on ogbn-Papers100M with 111M nodes and 1.6B edges.
| no_new_dataset | 0.9463 |
2304.04816 | Cheng-Yen Yang | Cheng-Yen Yang, Alan Yu Shyang Tan, Melanie J. Underwood, Charlotte
Bodie, Zhongyu Jiang, Steve George, Karl Warr, Jenq-Neng Hwang, Emma Jones | Multi-Object Tracking by Iteratively Associating Detections with Uniform
Appearance for Trawl-Based Fishing Bycatch Monitoring | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The aim of in-trawl catch monitoring for use in fishing operations is to
detect, track and classify fish targets in real-time from video footage.
Information gathered could be used to release unwanted bycatch in real-time.
However, traditional multi-object tracking (MOT) methods have limitations, as
they are developed for tracking vehicles or pedestrians with linear motions and
diverse appearances, which are different from the scenarios such as livestock
monitoring. Therefore, we propose a novel MOT method, built upon an existing
observation-centric tracking algorithm, by adopting a new iterative association
step to significantly boost the performance of tracking targets with a uniform
appearance. The iterative association module is designed as an extendable
component that can be merged into most existing tracking methods. Our method
offers improved performance in tracking targets with uniform appearance and
outperforms state-of-the-art techniques on our underwater fish datasets as well
as the MOT17 dataset, without increasing latency nor sacrificing accuracy as
measured by HOTA, MOTA, and IDF1 performance metrics.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 18:55:10 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Yang",
"Cheng-Yen",
""
],
[
"Tan",
"Alan Yu Shyang",
""
],
[
"Underwood",
"Melanie J.",
""
],
[
"Bodie",
"Charlotte",
""
],
[
"Jiang",
"Zhongyu",
""
],
[
"George",
"Steve",
""
],
[
"Warr",
"Karl",
""
],
[
"Hwang",
"Jenq-Neng",
""
],
[
"Jones",
"Emma",
""
]
] | TITLE: Multi-Object Tracking by Iteratively Associating Detections with Uniform
Appearance for Trawl-Based Fishing Bycatch Monitoring
ABSTRACT: The aim of in-trawl catch monitoring for use in fishing operations is to
detect, track and classify fish targets in real-time from video footage.
Information gathered could be used to release unwanted bycatch in real-time.
However, traditional multi-object tracking (MOT) methods have limitations, as
they are developed for tracking vehicles or pedestrians with linear motions and
diverse appearances, which are different from the scenarios such as livestock
monitoring. Therefore, we propose a novel MOT method, built upon an existing
observation-centric tracking algorithm, by adopting a new iterative association
step to significantly boost the performance of tracking targets with a uniform
appearance. The iterative association module is designed as an extendable
component that can be merged into most existing tracking methods. Our method
offers improved performance in tracking targets with uniform appearance and
outperforms state-of-the-art techniques on our underwater fish datasets as well
as the MOT17 dataset, without increasing latency nor sacrificing accuracy as
measured by HOTA, MOTA, and IDF1 performance metrics.
| no_new_dataset | 0.939526 |
2304.04817 | Thomas H\"utter | Konstantin Emil Thiel and Daniel Kocher and Nikolaus Augsten and
Thomas H\"utter and Willi Mann and Daniel Ulrich Schmitt | FINEX: A Fast Index for Exact & Flexible Density-Based Clustering
(Extended Version with Proofs)* | null | null | null | null | cs.DB | http://creativecommons.org/licenses/by/4.0/ | Density-based clustering aims to find groups of similar objects (i.e.,
clusters) in a given dataset. Applications include, e.g., process mining and
anomaly detection. It comes with two user parameters ({\epsilon}, MinPts) that
determine the clustering result, but are typically unknown in advance. Thus,
users need to interactively test various settings until satisfying clusterings
are found. However, existing solutions suffer from the following limitations:
(a) Ineffective pruning of expensive neighborhood computations. (b) Approximate
clustering, where objects are falsely labeled noise. (c) Restricted parameter
tuning that is limited to {\epsilon} whereas MinPts is constant, which reduces
the explorable clusterings. (d) Inflexibility in terms of applicable data types
and distance functions. We propose FINEX, a linear-space index that overcomes
these limitations. Our index provides exact clusterings and can be queried with
either of the two parameters. FINEX avoids neighborhood computations where
possible and reduces the complexities of the remaining computations by
leveraging fundamental properties of density-based clusters. Hence, our
solution is effcient and flexible regarding data types and distance functions.
Moreover, FINEX respects the original and straightforward notion of
density-based clustering. In our experiments on 12 large real-world datasets
from various domains, FINEX frequently outperforms state-of-the-art techniques
for exact clustering by orders of magnitude.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 18:57:45 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Thiel",
"Konstantin Emil",
""
],
[
"Kocher",
"Daniel",
""
],
[
"Augsten",
"Nikolaus",
""
],
[
"Hütter",
"Thomas",
""
],
[
"Mann",
"Willi",
""
],
[
"Schmitt",
"Daniel Ulrich",
""
]
] | TITLE: FINEX: A Fast Index for Exact & Flexible Density-Based Clustering
(Extended Version with Proofs)*
ABSTRACT: Density-based clustering aims to find groups of similar objects (i.e.,
clusters) in a given dataset. Applications include, e.g., process mining and
anomaly detection. It comes with two user parameters ({\epsilon}, MinPts) that
determine the clustering result, but are typically unknown in advance. Thus,
users need to interactively test various settings until satisfying clusterings
are found. However, existing solutions suffer from the following limitations:
(a) Ineffective pruning of expensive neighborhood computations. (b) Approximate
clustering, where objects are falsely labeled noise. (c) Restricted parameter
tuning that is limited to {\epsilon} whereas MinPts is constant, which reduces
the explorable clusterings. (d) Inflexibility in terms of applicable data types
and distance functions. We propose FINEX, a linear-space index that overcomes
these limitations. Our index provides exact clusterings and can be queried with
either of the two parameters. FINEX avoids neighborhood computations where
possible and reduces the complexities of the remaining computations by
leveraging fundamental properties of density-based clusters. Hence, our
solution is effcient and flexible regarding data types and distance functions.
Moreover, FINEX respects the original and straightforward notion of
density-based clustering. In our experiments on 12 large real-world datasets
from various domains, FINEX frequently outperforms state-of-the-art techniques
for exact clustering by orders of magnitude.
| no_new_dataset | 0.949153 |
2304.04820 | Ze Wang | Ze Wang, Jiang Wang, Zicheng Liu, and Qiang Qiu | Binary Latent Diffusion | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we show that a binary latent space can be explored for compact
yet expressive image representations. We model the bi-directional mappings
between an image and the corresponding latent binary representation by training
an auto-encoder with a Bernoulli encoding distribution. On the one hand, the
binary latent space provides a compact discrete image representation of which
the distribution can be modeled more efficiently than pixels or continuous
latent representations. On the other hand, we now represent each image patch as
a binary vector instead of an index of a learned cookbook as in discrete image
representations with vector quantization. In this way, we obtain binary latent
representations that allow for better image quality and high-resolution image
representations without any multi-stage hierarchy in the latent space. In this
binary latent space, images can now be generated effectively using a binary
latent diffusion model tailored specifically for modeling the prior over the
binary image representations. We present both conditional and unconditional
image generation experiments with multiple datasets, and show that the proposed
method performs comparably to state-of-the-art methods while dramatically
improving the sampling efficiency to as few as 16 steps without using any
test-time acceleration. The proposed framework can also be seamlessly scaled to
$1024 \times 1024$ high-resolution image generation without resorting to latent
hierarchy or multi-stage refinements.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 19:03:28 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wang",
"Ze",
""
],
[
"Wang",
"Jiang",
""
],
[
"Liu",
"Zicheng",
""
],
[
"Qiu",
"Qiang",
""
]
] | TITLE: Binary Latent Diffusion
ABSTRACT: In this paper, we show that a binary latent space can be explored for compact
yet expressive image representations. We model the bi-directional mappings
between an image and the corresponding latent binary representation by training
an auto-encoder with a Bernoulli encoding distribution. On the one hand, the
binary latent space provides a compact discrete image representation of which
the distribution can be modeled more efficiently than pixels or continuous
latent representations. On the other hand, we now represent each image patch as
a binary vector instead of an index of a learned cookbook as in discrete image
representations with vector quantization. In this way, we obtain binary latent
representations that allow for better image quality and high-resolution image
representations without any multi-stage hierarchy in the latent space. In this
binary latent space, images can now be generated effectively using a binary
latent diffusion model tailored specifically for modeling the prior over the
binary image representations. We present both conditional and unconditional
image generation experiments with multiple datasets, and show that the proposed
method performs comparably to state-of-the-art methods while dramatically
improving the sampling efficiency to as few as 16 steps without using any
test-time acceleration. The proposed framework can also be seamlessly scaled to
$1024 \times 1024$ high-resolution image generation without resorting to latent
hierarchy or multi-stage refinements.
| no_new_dataset | 0.951594 |
2304.04836 | Weiwei Sun | Weiwei Sun, Pengjie Ren, Zhaochun Ren | Generative Knowledge Selection for Knowledge-Grounded Dialogues | Findings of EACL-23 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Knowledge selection is the key in knowledge-grounded dialogues (KGD), which
aims to select an appropriate knowledge snippet to be used in the utterance
based on dialogue history. Previous studies mainly employ the classification
approach to classify each candidate snippet as "relevant" or "irrelevant"
independently. However, such approaches neglect the interactions between
snippets, leading to difficulties in inferring the meaning of snippets.
Moreover, they lack modeling of the discourse structure of dialogue-knowledge
interactions. We propose a simple yet effective generative approach for
knowledge selection, called GenKS. GenKS learns to select snippets by
generating their identifiers with a sequence-to-sequence model. GenKS therefore
captures intra-knowledge interaction inherently through attention mechanisms.
Meanwhile, we devise a hyperlink mechanism to model the dialogue-knowledge
interactions explicitly. We conduct experiments on three benchmark datasets,
and verify GenKS achieves the best results on both knowledge selection and
response generation.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 19:49:55 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Sun",
"Weiwei",
""
],
[
"Ren",
"Pengjie",
""
],
[
"Ren",
"Zhaochun",
""
]
] | TITLE: Generative Knowledge Selection for Knowledge-Grounded Dialogues
ABSTRACT: Knowledge selection is the key in knowledge-grounded dialogues (KGD), which
aims to select an appropriate knowledge snippet to be used in the utterance
based on dialogue history. Previous studies mainly employ the classification
approach to classify each candidate snippet as "relevant" or "irrelevant"
independently. However, such approaches neglect the interactions between
snippets, leading to difficulties in inferring the meaning of snippets.
Moreover, they lack modeling of the discourse structure of dialogue-knowledge
interactions. We propose a simple yet effective generative approach for
knowledge selection, called GenKS. GenKS learns to select snippets by
generating their identifiers with a sequence-to-sequence model. GenKS therefore
captures intra-knowledge interaction inherently through attention mechanisms.
Meanwhile, we devise a hyperlink mechanism to model the dialogue-knowledge
interactions explicitly. We conduct experiments on three benchmark datasets,
and verify GenKS achieves the best results on both knowledge selection and
response generation.
| no_new_dataset | 0.946547 |
2304.04848 | Wenqi Xian | Wenqi Xian and Alja\v{z} Bo\v{z}i\v{c} and Noah Snavely and Christoph
Lassner | Neural Lens Modeling | To be presented at CVPR 2023, Project webpage:
https://neural-lens.github.io | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Recent methods for 3D reconstruction and rendering increasingly benefit from
end-to-end optimization of the entire image formation process. However, this
approach is currently limited: effects of the optical hardware stack and in
particular lenses are hard to model in a unified way. This limits the quality
that can be achieved for camera calibration and the fidelity of the results of
3D reconstruction. In this paper, we propose NeuroLens, a neural lens model for
distortion and vignetting that can be used for point projection and ray casting
and can be optimized through both operations. This means that it can
(optionally) be used to perform pre-capture calibration using classical
calibration targets, and can later be used to perform calibration or refinement
during 3D reconstruction, e.g., while optimizing a radiance field. To evaluate
the performance of our proposed model, we create a comprehensive dataset
assembled from the Lensfun database with a multitude of lenses. Using this and
other real-world datasets, we show that the quality of our proposed lens model
outperforms standard packages as well as recent approaches while being much
easier to use and extend. The model generalizes across many lens types and is
trivial to integrate into existing 3D reconstruction and rendering systems.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 20:09:17 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Xian",
"Wenqi",
""
],
[
"Božič",
"Aljaž",
""
],
[
"Snavely",
"Noah",
""
],
[
"Lassner",
"Christoph",
""
]
] | TITLE: Neural Lens Modeling
ABSTRACT: Recent methods for 3D reconstruction and rendering increasingly benefit from
end-to-end optimization of the entire image formation process. However, this
approach is currently limited: effects of the optical hardware stack and in
particular lenses are hard to model in a unified way. This limits the quality
that can be achieved for camera calibration and the fidelity of the results of
3D reconstruction. In this paper, we propose NeuroLens, a neural lens model for
distortion and vignetting that can be used for point projection and ray casting
and can be optimized through both operations. This means that it can
(optionally) be used to perform pre-capture calibration using classical
calibration targets, and can later be used to perform calibration or refinement
during 3D reconstruction, e.g., while optimizing a radiance field. To evaluate
the performance of our proposed model, we create a comprehensive dataset
assembled from the Lensfun database with a multitude of lenses. Using this and
other real-world datasets, we show that the quality of our proposed lens model
outperforms standard packages as well as recent approaches while being much
easier to use and extend. The model generalizes across many lens types and is
trivial to integrate into existing 3D reconstruction and rendering systems.
| new_dataset | 0.959383 |
2304.04858 | Amirmohammad Sarfi | Amirmohammad Sarfi, Zahra Karimpour, Muawiz Chaudhary, Nasir M.
Khalid, Mirco Ravanelli, Sudhir Mudur and Eugene Belilovsky | Simulated Annealing in Early Layers Leads to Better Generalization | null | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, a number of iterative learning methods have been introduced to
improve generalization. These typically rely on training for longer periods of
time in exchange for improved generalization. LLF (later-layer-forgetting) is a
state-of-the-art method in this category. It strengthens learning in early
layers by periodically re-initializing the last few layers of the network. Our
principal innovation in this work is to use Simulated annealing in EArly Layers
(SEAL) of the network in place of re-initialization of later layers.
Essentially, later layers go through the normal gradient descent process, while
the early layers go through short stints of gradient ascent followed by
gradient descent. Extensive experiments on the popular Tiny-ImageNet dataset
benchmark and a series of transfer learning and few-shot learning tasks show
that we outperform LLF by a significant margin. We further show that, compared
to normal training, LLF features, although improving on the target task,
degrade the transfer learning performance across all datasets we explored. In
comparison, our method outperforms LLF across the same target datasets by a
large margin. We also show that the prediction depth of our method is
significantly lower than that of LLF and normal training, indicating on average
better prediction performance.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 20:41:40 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Sarfi",
"Amirmohammad",
""
],
[
"Karimpour",
"Zahra",
""
],
[
"Chaudhary",
"Muawiz",
""
],
[
"Khalid",
"Nasir M.",
""
],
[
"Ravanelli",
"Mirco",
""
],
[
"Mudur",
"Sudhir",
""
],
[
"Belilovsky",
"Eugene",
""
]
] | TITLE: Simulated Annealing in Early Layers Leads to Better Generalization
ABSTRACT: Recently, a number of iterative learning methods have been introduced to
improve generalization. These typically rely on training for longer periods of
time in exchange for improved generalization. LLF (later-layer-forgetting) is a
state-of-the-art method in this category. It strengthens learning in early
layers by periodically re-initializing the last few layers of the network. Our
principal innovation in this work is to use Simulated annealing in EArly Layers
(SEAL) of the network in place of re-initialization of later layers.
Essentially, later layers go through the normal gradient descent process, while
the early layers go through short stints of gradient ascent followed by
gradient descent. Extensive experiments on the popular Tiny-ImageNet dataset
benchmark and a series of transfer learning and few-shot learning tasks show
that we outperform LLF by a significant margin. We further show that, compared
to normal training, LLF features, although improving on the target task,
degrade the transfer learning performance across all datasets we explored. In
comparison, our method outperforms LLF across the same target datasets by a
large margin. We also show that the prediction depth of our method is
significantly lower than that of LLF and normal training, indicating on average
better prediction performance.
| no_new_dataset | 0.951051 |
2304.04875 | Gyeongsik Moon | Gyeongsik Moon, Hongsuk Choi, Sanghyuk Chun, Jiyoung Lee, Sangdoo Yun | Three Recipes for Better 3D Pseudo-GTs of 3D Human Mesh Estimation in
the Wild | Published at CVPRW 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Recovering 3D human mesh in the wild is greatly challenging as in-the-wild
(ITW) datasets provide only 2D pose ground truths (GTs). Recently, 3D
pseudo-GTs have been widely used to train 3D human mesh estimation networks as
the 3D pseudo-GTs enable 3D mesh supervision when training the networks on ITW
datasets. However, despite the great potential of the 3D pseudo-GTs, there has
been no extensive analysis that investigates which factors are important to
make more beneficial 3D pseudo-GTs. In this paper, we provide three recipes to
obtain highly beneficial 3D pseudo-GTs of ITW datasets. The main challenge is
that only 2D-based weak supervision is allowed when obtaining the 3D
pseudo-GTs. Each of our three recipes addresses the challenge in each aspect:
depth ambiguity, sub-optimality of weak supervision, and implausible
articulation. Experimental results show that simply re-training
state-of-the-art networks with our new 3D pseudo-GTs elevates their performance
to the next level without bells and whistles. The 3D pseudo-GT is publicly
available in https://github.com/mks0601/NeuralAnnot_RELEASE.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 21:43:02 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Moon",
"Gyeongsik",
""
],
[
"Choi",
"Hongsuk",
""
],
[
"Chun",
"Sanghyuk",
""
],
[
"Lee",
"Jiyoung",
""
],
[
"Yun",
"Sangdoo",
""
]
] | TITLE: Three Recipes for Better 3D Pseudo-GTs of 3D Human Mesh Estimation in
the Wild
ABSTRACT: Recovering 3D human mesh in the wild is greatly challenging as in-the-wild
(ITW) datasets provide only 2D pose ground truths (GTs). Recently, 3D
pseudo-GTs have been widely used to train 3D human mesh estimation networks as
the 3D pseudo-GTs enable 3D mesh supervision when training the networks on ITW
datasets. However, despite the great potential of the 3D pseudo-GTs, there has
been no extensive analysis that investigates which factors are important to
make more beneficial 3D pseudo-GTs. In this paper, we provide three recipes to
obtain highly beneficial 3D pseudo-GTs of ITW datasets. The main challenge is
that only 2D-based weak supervision is allowed when obtaining the 3D
pseudo-GTs. Each of our three recipes addresses the challenge in each aspect:
depth ambiguity, sub-optimality of weak supervision, and implausible
articulation. Experimental results show that simply re-training
state-of-the-art networks with our new 3D pseudo-GTs elevates their performance
to the next level without bells and whistles. The 3D pseudo-GT is publicly
available in https://github.com/mks0601/NeuralAnnot_RELEASE.
| no_new_dataset | 0.945751 |
2304.04897 | Youngjoong Kwon | Youngjoong Kwon, Dahun Kim, Duygu Ceylan, Henry Fuchs | Neural Image-based Avatars: Generalizable Radiance Fields for Human
Avatar Modeling | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We present a method that enables synthesizing novel views and novel poses of
arbitrary human performers from sparse multi-view images. A key ingredient of
our method is a hybrid appearance blending module that combines the advantages
of the implicit body NeRF representation and image-based rendering. Existing
generalizable human NeRF methods that are conditioned on the body model have
shown robustness against the geometric variation of arbitrary human performers.
Yet they often exhibit blurry results when generalized onto unseen identities.
Meanwhile, image-based rendering shows high-quality results when sufficient
observations are available, whereas it suffers artifacts in sparse-view
settings. We propose Neural Image-based Avatars (NIA) that exploits the best of
those two methods: to maintain robustness under new articulations and
self-occlusions while directly leveraging the available (sparse) source view
colors to preserve appearance details of new subject identities. Our hybrid
design outperforms recent methods on both in-domain identity generalization as
well as challenging cross-dataset generalization settings. Also, in terms of
the pose generalization, our method outperforms even the per-subject optimized
animatable NeRF methods. The video results are available at
https://youngjoongunc.github.io/nia
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 23:53:28 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Kwon",
"Youngjoong",
""
],
[
"Kim",
"Dahun",
""
],
[
"Ceylan",
"Duygu",
""
],
[
"Fuchs",
"Henry",
""
]
] | TITLE: Neural Image-based Avatars: Generalizable Radiance Fields for Human
Avatar Modeling
ABSTRACT: We present a method that enables synthesizing novel views and novel poses of
arbitrary human performers from sparse multi-view images. A key ingredient of
our method is a hybrid appearance blending module that combines the advantages
of the implicit body NeRF representation and image-based rendering. Existing
generalizable human NeRF methods that are conditioned on the body model have
shown robustness against the geometric variation of arbitrary human performers.
Yet they often exhibit blurry results when generalized onto unseen identities.
Meanwhile, image-based rendering shows high-quality results when sufficient
observations are available, whereas it suffers artifacts in sparse-view
settings. We propose Neural Image-based Avatars (NIA) that exploits the best of
those two methods: to maintain robustness under new articulations and
self-occlusions while directly leveraging the available (sparse) source view
colors to preserve appearance details of new subject identities. Our hybrid
design outperforms recent methods on both in-domain identity generalization as
well as challenging cross-dataset generalization settings. Also, in terms of
the pose generalization, our method outperforms even the per-subject optimized
animatable NeRF methods. The video results are available at
https://youngjoongunc.github.io/nia
| no_new_dataset | 0.951323 |
2304.04907 | Jialu Li | Jialu Li, Mohit Bansal | Improving Vision-and-Language Navigation by Generating Future-View Image
Semantics | CVPR 2023 (Project webpage: https://jialuli-luka.github.io/VLN-SIG) | null | null | null | cs.CV cs.AI cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Vision-and-Language Navigation (VLN) is the task that requires an agent to
navigate through the environment based on natural language instructions. At
each step, the agent takes the next action by selecting from a set of navigable
locations. In this paper, we aim to take one step further and explore whether
the agent can benefit from generating the potential future view during
navigation. Intuitively, humans will have an expectation of how the future
environment will look like, based on the natural language instructions and
surrounding views, which will aid correct navigation. Hence, to equip the agent
with this ability to generate the semantics of future navigation views, we
first propose three proxy tasks during the agent's in-domain pre-training:
Masked Panorama Modeling (MPM), Masked Trajectory Modeling (MTM), and Action
Prediction with Image Generation (APIG). These three objectives teach the model
to predict missing views in a panorama (MPM), predict missing steps in the full
trajectory (MTM), and generate the next view based on the full instruction and
navigation history (APIG), respectively. We then fine-tune the agent on the VLN
task with an auxiliary loss that minimizes the difference between the view
semantics generated by the agent and the ground truth view semantics of the
next step. Empirically, our VLN-SIG achieves the new state-of-the-art on both
the Room-to-Room dataset and the CVDN dataset. We further show that our agent
learns to fill in missing patches in future views qualitatively, which brings
more interpretability over agents' predicted actions. Lastly, we demonstrate
that learning to predict future view semantics also enables the agent to have
better performance on longer paths.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 00:36:02 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Li",
"Jialu",
""
],
[
"Bansal",
"Mohit",
""
]
] | TITLE: Improving Vision-and-Language Navigation by Generating Future-View Image
Semantics
ABSTRACT: Vision-and-Language Navigation (VLN) is the task that requires an agent to
navigate through the environment based on natural language instructions. At
each step, the agent takes the next action by selecting from a set of navigable
locations. In this paper, we aim to take one step further and explore whether
the agent can benefit from generating the potential future view during
navigation. Intuitively, humans will have an expectation of how the future
environment will look like, based on the natural language instructions and
surrounding views, which will aid correct navigation. Hence, to equip the agent
with this ability to generate the semantics of future navigation views, we
first propose three proxy tasks during the agent's in-domain pre-training:
Masked Panorama Modeling (MPM), Masked Trajectory Modeling (MTM), and Action
Prediction with Image Generation (APIG). These three objectives teach the model
to predict missing views in a panorama (MPM), predict missing steps in the full
trajectory (MTM), and generate the next view based on the full instruction and
navigation history (APIG), respectively. We then fine-tune the agent on the VLN
task with an auxiliary loss that minimizes the difference between the view
semantics generated by the agent and the ground truth view semantics of the
next step. Empirically, our VLN-SIG achieves the new state-of-the-art on both
the Room-to-Room dataset and the CVDN dataset. We further show that our agent
learns to fill in missing patches in future views qualitatively, which brings
more interpretability over agents' predicted actions. Lastly, we demonstrate
that learning to predict future view semantics also enables the agent to have
better performance on longer paths.
| no_new_dataset | 0.939359 |
2304.04917 | Xianrui Luo | Xianrui Luo, Juewen Peng, Weiyue Zhao, Ke Xian, Hao Lu, and Zhiguo Cao | Point-and-Shoot All-in-Focus Photo Synthesis from Smartphone Camera Pair | Early Access by IEEE Transactions on Circuits and Systems for Video
Technology 2022 | null | 10.1109/TCSVT.2022.3222609 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | All-in-Focus (AIF) photography is expected to be a commercial selling point
for modern smartphones. Standard AIF synthesis requires manual, time-consuming
operations such as focal stack compositing, which is unfriendly to ordinary
people. To achieve point-and-shoot AIF photography with a smartphone, we expect
that an AIF photo can be generated from one shot of the scene, instead of from
multiple photos captured by the same camera. Benefiting from the multi-camera
module in modern smartphones, we introduce a new task of AIF synthesis from
main (wide) and ultra-wide cameras. The goal is to recover sharp details from
defocused regions in the main-camera photo with the help of the
ultra-wide-camera one. The camera setting poses new challenges such as
parallax-induced occlusions and inconsistent color between cameras. To overcome
the challenges, we introduce a predict-and-refine network to mitigate
occlusions and propose dynamic frequency-domain alignment for color correction.
To enable effective training and evaluation, we also build an AIF dataset with
2686 unique scenes. Each scene includes two photos captured by the main camera,
one photo captured by the ultrawide camera, and a synthesized AIF photo.
Results show that our solution, termed EasyAIF, can produce high-quality AIF
photos and outperforms strong baselines quantitatively and qualitatively. For
the first time, we demonstrate point-and-shoot AIF photo synthesis successfully
from main and ultra-wide cameras.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 01:09:54 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Luo",
"Xianrui",
""
],
[
"Peng",
"Juewen",
""
],
[
"Zhao",
"Weiyue",
""
],
[
"Xian",
"Ke",
""
],
[
"Lu",
"Hao",
""
],
[
"Cao",
"Zhiguo",
""
]
] | TITLE: Point-and-Shoot All-in-Focus Photo Synthesis from Smartphone Camera Pair
ABSTRACT: All-in-Focus (AIF) photography is expected to be a commercial selling point
for modern smartphones. Standard AIF synthesis requires manual, time-consuming
operations such as focal stack compositing, which is unfriendly to ordinary
people. To achieve point-and-shoot AIF photography with a smartphone, we expect
that an AIF photo can be generated from one shot of the scene, instead of from
multiple photos captured by the same camera. Benefiting from the multi-camera
module in modern smartphones, we introduce a new task of AIF synthesis from
main (wide) and ultra-wide cameras. The goal is to recover sharp details from
defocused regions in the main-camera photo with the help of the
ultra-wide-camera one. The camera setting poses new challenges such as
parallax-induced occlusions and inconsistent color between cameras. To overcome
the challenges, we introduce a predict-and-refine network to mitigate
occlusions and propose dynamic frequency-domain alignment for color correction.
To enable effective training and evaluation, we also build an AIF dataset with
2686 unique scenes. Each scene includes two photos captured by the main camera,
one photo captured by the ultrawide camera, and a synthesized AIF photo.
Results show that our solution, termed EasyAIF, can produce high-quality AIF
photos and outperforms strong baselines quantitatively and qualitatively. For
the first time, we demonstrate point-and-shoot AIF photo synthesis successfully
from main and ultra-wide cameras.
| new_dataset | 0.956391 |
2304.04952 | Guanyi Qin | Guanyi Qin, Runze Hu, Yutao Liu, Xiawu Zheng, Haotian Liu, Xiu Li, Yan
Zhang | Data-Efficient Image Quality Assessment with Attention-Panel Decoder | Accepted by AAAI 2023 | null | null | null | cs.CV cs.LG eess.IV | http://creativecommons.org/licenses/by/4.0/ | Blind Image Quality Assessment (BIQA) is a fundamental task in computer
vision, which however remains unresolved due to the complex distortion
conditions and diversified image contents. To confront this challenge, we in
this paper propose a novel BIQA pipeline based on the Transformer architecture,
which achieves an efficient quality-aware feature representation with much
fewer data. More specifically, we consider the traditional fine-tuning in BIQA
as an interpretation of the pre-trained model. In this way, we further
introduce a Transformer decoder to refine the perceptual information of the CLS
token from different perspectives. This enables our model to establish the
quality-aware feature manifold efficiently while attaining a strong
generalization capability. Meanwhile, inspired by the subjective evaluation
behaviors of human, we introduce a novel attention panel mechanism, which
improves the model performance and reduces the prediction uncertainty
simultaneously. The proposed BIQA method maintains a lightweight design with
only one layer of the decoder, yet extensive experiments on eight standard BIQA
datasets (both synthetic and authentic) demonstrate its superior performance to
the state-of-the-art BIQA methods, i.e., achieving the SRCC values of 0.875
(vs. 0.859 in LIVEC) and 0.980 (vs. 0.969 in LIVE).
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 03:52:17 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Qin",
"Guanyi",
""
],
[
"Hu",
"Runze",
""
],
[
"Liu",
"Yutao",
""
],
[
"Zheng",
"Xiawu",
""
],
[
"Liu",
"Haotian",
""
],
[
"Li",
"Xiu",
""
],
[
"Zhang",
"Yan",
""
]
] | TITLE: Data-Efficient Image Quality Assessment with Attention-Panel Decoder
ABSTRACT: Blind Image Quality Assessment (BIQA) is a fundamental task in computer
vision, which however remains unresolved due to the complex distortion
conditions and diversified image contents. To confront this challenge, we in
this paper propose a novel BIQA pipeline based on the Transformer architecture,
which achieves an efficient quality-aware feature representation with much
fewer data. More specifically, we consider the traditional fine-tuning in BIQA
as an interpretation of the pre-trained model. In this way, we further
introduce a Transformer decoder to refine the perceptual information of the CLS
token from different perspectives. This enables our model to establish the
quality-aware feature manifold efficiently while attaining a strong
generalization capability. Meanwhile, inspired by the subjective evaluation
behaviors of human, we introduce a novel attention panel mechanism, which
improves the model performance and reduces the prediction uncertainty
simultaneously. The proposed BIQA method maintains a lightweight design with
only one layer of the decoder, yet extensive experiments on eight standard BIQA
datasets (both synthetic and authentic) demonstrate its superior performance to
the state-of-the-art BIQA methods, i.e., achieving the SRCC values of 0.875
(vs. 0.859 in LIVEC) and 0.980 (vs. 0.969 in LIVE).
| no_new_dataset | 0.941654 |
2304.04956 | Hong Wei Ren | Hongwei Ren, Yuhong Shi, Kewei Liang | Multi-Graph Convolution Network for Pose Forecasting | arXiv admin note: text overlap with arXiv:2110.04573 by other authors | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, there has been a growing interest in predicting human motion, which
involves forecasting future body poses based on observed pose sequences. This
task is complex due to modeling spatial and temporal relationships. The most
commonly used models for this task are autoregressive models, such as recurrent
neural networks (RNNs) or variants, and Transformer Networks. However, RNNs
have several drawbacks, such as vanishing or exploding gradients. Other
researchers have attempted to solve the communication problem in the spatial
dimension by integrating Graph Convolutional Networks (GCN) and Long Short-Term
Memory (LSTM) models. These works deal with temporal and spatial information
separately, which limits the effectiveness. To fix this problem, we propose a
novel approach called the multi-graph convolution network (MGCN) for 3D human
pose forecasting. This model simultaneously captures spatial and temporal
information by introducing an augmented graph for pose sequences. Multiple
frames give multiple parts, joined together in a single graph instance.
Furthermore, we also explore the influence of natural structure and
sequence-aware attention to our model. In our experimental evaluation of the
large-scale benchmark datasets, Human3.6M, AMSS and 3DPW, MGCN outperforms the
state-of-the-art in pose prediction.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 03:59:43 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Ren",
"Hongwei",
""
],
[
"Shi",
"Yuhong",
""
],
[
"Liang",
"Kewei",
""
]
] | TITLE: Multi-Graph Convolution Network for Pose Forecasting
ABSTRACT: Recently, there has been a growing interest in predicting human motion, which
involves forecasting future body poses based on observed pose sequences. This
task is complex due to modeling spatial and temporal relationships. The most
commonly used models for this task are autoregressive models, such as recurrent
neural networks (RNNs) or variants, and Transformer Networks. However, RNNs
have several drawbacks, such as vanishing or exploding gradients. Other
researchers have attempted to solve the communication problem in the spatial
dimension by integrating Graph Convolutional Networks (GCN) and Long Short-Term
Memory (LSTM) models. These works deal with temporal and spatial information
separately, which limits the effectiveness. To fix this problem, we propose a
novel approach called the multi-graph convolution network (MGCN) for 3D human
pose forecasting. This model simultaneously captures spatial and temporal
information by introducing an augmented graph for pose sequences. Multiple
frames give multiple parts, joined together in a single graph instance.
Furthermore, we also explore the influence of natural structure and
sequence-aware attention to our model. In our experimental evaluation of the
large-scale benchmark datasets, Human3.6M, AMSS and 3DPW, MGCN outperforms the
state-of-the-art in pose prediction.
| no_new_dataset | 0.941385 |
2304.04960 | Soohyun Kim | Soohyun Kim, Junho Kim, Taekyung Kim, Hwan Heo, Seungryong Kim,
Jiyoung Lee, Jin-Hwa Kim | Panoramic Image-to-Image Translation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | In this paper, we tackle the challenging task of Panoramic Image-to-Image
translation (Pano-I2I) for the first time. This task is difficult due to the
geometric distortion of panoramic images and the lack of a panoramic image
dataset with diverse conditions, like weather or time. To address these
challenges, we propose a panoramic distortion-aware I2I model that preserves
the structure of the panoramic images while consistently translating their
global style referenced from a pinhole image. To mitigate the distortion issue
in naive 360 panorama translation, we adopt spherical positional embedding to
our transformer encoders, introduce a distortion-free discriminator, and apply
sphere-based rotation for augmentation and its ensemble. We also design a
content encoder and a style encoder to be deformation-aware to deal with a
large domain gap between panoramas and pinhole images, enabling us to work on
diverse conditions of pinhole images. In addition, considering the large
discrepancy between panoramas and pinhole images, our framework decouples the
learning procedure of the panoramic reconstruction stage from the translation
stage. We show distinct improvements over existing I2I models in translating
the StreetLearn dataset in the daytime into diverse conditions. The code will
be publicly available online for our community.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 04:08:58 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Kim",
"Soohyun",
""
],
[
"Kim",
"Junho",
""
],
[
"Kim",
"Taekyung",
""
],
[
"Heo",
"Hwan",
""
],
[
"Kim",
"Seungryong",
""
],
[
"Lee",
"Jiyoung",
""
],
[
"Kim",
"Jin-Hwa",
""
]
] | TITLE: Panoramic Image-to-Image Translation
ABSTRACT: In this paper, we tackle the challenging task of Panoramic Image-to-Image
translation (Pano-I2I) for the first time. This task is difficult due to the
geometric distortion of panoramic images and the lack of a panoramic image
dataset with diverse conditions, like weather or time. To address these
challenges, we propose a panoramic distortion-aware I2I model that preserves
the structure of the panoramic images while consistently translating their
global style referenced from a pinhole image. To mitigate the distortion issue
in naive 360 panorama translation, we adopt spherical positional embedding to
our transformer encoders, introduce a distortion-free discriminator, and apply
sphere-based rotation for augmentation and its ensemble. We also design a
content encoder and a style encoder to be deformation-aware to deal with a
large domain gap between panoramas and pinhole images, enabling us to work on
diverse conditions of pinhole images. In addition, considering the large
discrepancy between panoramas and pinhole images, our framework decouples the
learning procedure of the panoramic reconstruction stage from the translation
stage. We show distinct improvements over existing I2I models in translating
the StreetLearn dataset in the daytime into diverse conditions. The code will
be publicly available online for our community.
| no_new_dataset | 0.943138 |
2304.04972 | Yunheng Shen | Yunheng Shen, Haoxiang Wang, Hairong Lv | Federated Learning with Classifier Shift for Class Imbalance | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Federated learning aims to learn a global model collaboratively while the
training data belongs to different clients and is not allowed to be exchanged.
However, the statistical heterogeneity challenge on non-IID data, such as class
imbalance in classification, will cause client drift and significantly reduce
the performance of the global model. This paper proposes a simple and effective
approach named FedShift which adds the shift on the classifier output during
the local training phase to alleviate the negative impact of class imbalance.
We theoretically prove that the classifier shift in FedShift can make the local
optimum consistent with the global optimum and ensure the convergence of the
algorithm. Moreover, our experiments indicate that FedShift significantly
outperforms the other state-of-the-art federated learning approaches on various
datasets regarding accuracy and communication efficiency.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 04:38:39 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Shen",
"Yunheng",
""
],
[
"Wang",
"Haoxiang",
""
],
[
"Lv",
"Hairong",
""
]
] | TITLE: Federated Learning with Classifier Shift for Class Imbalance
ABSTRACT: Federated learning aims to learn a global model collaboratively while the
training data belongs to different clients and is not allowed to be exchanged.
However, the statistical heterogeneity challenge on non-IID data, such as class
imbalance in classification, will cause client drift and significantly reduce
the performance of the global model. This paper proposes a simple and effective
approach named FedShift which adds the shift on the classifier output during
the local training phase to alleviate the negative impact of class imbalance.
We theoretically prove that the classifier shift in FedShift can make the local
optimum consistent with the global optimum and ensure the convergence of the
algorithm. Moreover, our experiments indicate that FedShift significantly
outperforms the other state-of-the-art federated learning approaches on various
datasets regarding accuracy and communication efficiency.
| no_new_dataset | 0.951142 |
2304.04983 | Akalanka Galappaththi | Akalanka Galappaththi, Sarah Nadi | A Data Set of Generalizable Python Code Change Patterns | null | null | null | null | cs.SE | http://creativecommons.org/licenses/by-sa/4.0/ | Mining repetitive code changes from version control history is a common way
of discovering unknown change patterns. Such change patterns can be used in
code recommender systems or automated program repair techniques. While there
are such tools and datasets exist for Java, there is little work on finding and
recommending such changes in Python. In this paper, we present a data set of
manually vetted generalizable Python repetitive code change patterns. We create
a coding guideline to identify generalizable change patterns that can be used
in automated tooling. We leverage the mined change patterns from recent work
that mines repetitive changes in Python projects and use our coding guideline
to manually review the patterns. For each change, we also record a description
of the change and why it is applied along with other characteristics such as
the number of projects it occurs in. This review process allows us to identify
and share 72 Python change patterns that can be used to build and advance
Python developer support tools.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 04:59:26 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Galappaththi",
"Akalanka",
""
],
[
"Nadi",
"Sarah",
""
]
] | TITLE: A Data Set of Generalizable Python Code Change Patterns
ABSTRACT: Mining repetitive code changes from version control history is a common way
of discovering unknown change patterns. Such change patterns can be used in
code recommender systems or automated program repair techniques. While there
are such tools and datasets exist for Java, there is little work on finding and
recommending such changes in Python. In this paper, we present a data set of
manually vetted generalizable Python repetitive code change patterns. We create
a coding guideline to identify generalizable change patterns that can be used
in automated tooling. We leverage the mined change patterns from recent work
that mines repetitive changes in Python projects and use our coding guideline
to manually review the patterns. For each change, we also record a description
of the change and why it is applied along with other characteristics such as
the number of projects it occurs in. This review process allows us to identify
and share 72 Python change patterns that can be used to build and advance
Python developer support tools.
| new_dataset | 0.601594 |
2304.04987 | Hassan Habibi Gharakheili | Ayyoob Hamza and Hassan Habibi Gharakheili and Theophilus A. Benson
and Gustavo Batista and Vijay Sivaraman | Detecting Anomalous Microflows in IoT Volumetric Attacks via Dynamic
Monitoring of MUD Activity | 18 pages, 13 figures | null | null | null | cs.CR cs.LG cs.NI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | IoT networks are increasingly becoming target of sophisticated new
cyber-attacks. Anomaly-based detection methods are promising in finding new
attacks, but there are certain practical challenges like false-positive alarms,
hard to explain, and difficult to scale cost-effectively. The IETF recent
standard called Manufacturer Usage Description (MUD) seems promising to limit
the attack surface on IoT devices by formally specifying their intended network
behavior. In this paper, we use SDN to enforce and monitor the expected
behaviors of each IoT device, and train one-class classifier models to detect
volumetric attacks.
Our specific contributions are fourfold. (1) We develop a multi-level
inferencing model to dynamically detect anomalous patterns in network activity
of MUD-compliant traffic flows via SDN telemetry, followed by packet inspection
of anomalous flows. This provides enhanced fine-grained visibility into
distributed and direct attacks, allowing us to precisely isolate volumetric
attacks with microflow (5-tuple) resolution. (2) We collect traffic traces
(benign and a variety of volumetric attacks) from network behavior of IoT
devices in our lab, generate labeled datasets, and make them available to the
public. (3) We prototype a full working system (modules are released as
open-source), demonstrates its efficacy in detecting volumetric attacks on
several consumer IoT devices with high accuracy while maintaining low false
positives, and provides insights into cost and performance of our system. (4)
We demonstrate how our models scale in environments with a large number of
connected IoTs (with datasets collected from a network of IP cameras in our
university campus) by considering various training strategies (per device unit
versus per device type), and balancing the accuracy of prediction against the
cost of models in terms of size and training time.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 05:17:51 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Hamza",
"Ayyoob",
""
],
[
"Gharakheili",
"Hassan Habibi",
""
],
[
"Benson",
"Theophilus A.",
""
],
[
"Batista",
"Gustavo",
""
],
[
"Sivaraman",
"Vijay",
""
]
] | TITLE: Detecting Anomalous Microflows in IoT Volumetric Attacks via Dynamic
Monitoring of MUD Activity
ABSTRACT: IoT networks are increasingly becoming target of sophisticated new
cyber-attacks. Anomaly-based detection methods are promising in finding new
attacks, but there are certain practical challenges like false-positive alarms,
hard to explain, and difficult to scale cost-effectively. The IETF recent
standard called Manufacturer Usage Description (MUD) seems promising to limit
the attack surface on IoT devices by formally specifying their intended network
behavior. In this paper, we use SDN to enforce and monitor the expected
behaviors of each IoT device, and train one-class classifier models to detect
volumetric attacks.
Our specific contributions are fourfold. (1) We develop a multi-level
inferencing model to dynamically detect anomalous patterns in network activity
of MUD-compliant traffic flows via SDN telemetry, followed by packet inspection
of anomalous flows. This provides enhanced fine-grained visibility into
distributed and direct attacks, allowing us to precisely isolate volumetric
attacks with microflow (5-tuple) resolution. (2) We collect traffic traces
(benign and a variety of volumetric attacks) from network behavior of IoT
devices in our lab, generate labeled datasets, and make them available to the
public. (3) We prototype a full working system (modules are released as
open-source), demonstrates its efficacy in detecting volumetric attacks on
several consumer IoT devices with high accuracy while maintaining low false
positives, and provides insights into cost and performance of our system. (4)
We demonstrate how our models scale in environments with a large number of
connected IoTs (with datasets collected from a network of IP cameras in our
university campus) by considering various training strategies (per device unit
versus per device type), and balancing the accuracy of prediction against the
cost of models in terms of size and training time.
| no_new_dataset | 0.934155 |
2304.04991 | Guangyong Wei | Guangyong Wei, Zhikui Duan, Shiren Li, Guangguang Yang, Xinmei Yu,
Junhua Li | Sim-T: Simplify the Transformer Network by Multiplexing Technique for
Speech Recognition | null | null | null | null | cs.SD cs.CL eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, a great deal of attention has been paid to the Transformer
network for speech recognition tasks due to its excellent model performance.
However, the Transformer network always involves heavy computation and large
number of parameters, causing serious deployment problems in devices with
limited computation sources or storage memory. In this paper, a new lightweight
model called Sim-T has been proposed to expand the generality of the
Transformer model. Under the help of the newly developed multiplexing
technique, the Sim-T can efficiently compress the model with negligible
sacrifice on its performance. To be more precise, the proposed technique
includes two parts, that are, module weight multiplexing and attention score
multiplexing. Moreover, a novel decoder structure has been proposed to
facilitate the attention score multiplexing. Extensive experiments have been
conducted to validate the effectiveness of Sim-T. In Aishell-1 dataset, when
the proposed Sim-T is 48% parameter less than the baseline Transformer, 0.4%
CER improvement can be obtained. Alternatively, 69% parameter reduction can be
achieved if the Sim-T gives the same performance as the baseline Transformer.
With regard to the HKUST and WSJ eval92 datasets, CER and WER will be improved
by 0.3% and 0.2%, respectively, when parameters in Sim-T are 40% less than the
baseline Transformer.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 05:25:00 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wei",
"Guangyong",
""
],
[
"Duan",
"Zhikui",
""
],
[
"Li",
"Shiren",
""
],
[
"Yang",
"Guangguang",
""
],
[
"Yu",
"Xinmei",
""
],
[
"Li",
"Junhua",
""
]
] | TITLE: Sim-T: Simplify the Transformer Network by Multiplexing Technique for
Speech Recognition
ABSTRACT: In recent years, a great deal of attention has been paid to the Transformer
network for speech recognition tasks due to its excellent model performance.
However, the Transformer network always involves heavy computation and large
number of parameters, causing serious deployment problems in devices with
limited computation sources or storage memory. In this paper, a new lightweight
model called Sim-T has been proposed to expand the generality of the
Transformer model. Under the help of the newly developed multiplexing
technique, the Sim-T can efficiently compress the model with negligible
sacrifice on its performance. To be more precise, the proposed technique
includes two parts, that are, module weight multiplexing and attention score
multiplexing. Moreover, a novel decoder structure has been proposed to
facilitate the attention score multiplexing. Extensive experiments have been
conducted to validate the effectiveness of Sim-T. In Aishell-1 dataset, when
the proposed Sim-T is 48% parameter less than the baseline Transformer, 0.4%
CER improvement can be obtained. Alternatively, 69% parameter reduction can be
achieved if the Sim-T gives the same performance as the baseline Transformer.
With regard to the HKUST and WSJ eval92 datasets, CER and WER will be improved
by 0.3% and 0.2%, respectively, when parameters in Sim-T are 40% less than the
baseline Transformer.
| no_new_dataset | 0.944587 |
2304.04994 | Boxin Du | Boxin Du, Lihui Liu, Jiejun Xu, Fei Wang, Hanghang Tong | Neural Multi-network Diffusion towards Social Recommendation | null | null | null | null | cs.LG cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph Neural Networks (GNNs) have been widely applied on a variety of
real-world applications, such as social recommendation. However, existing
GNN-based models on social recommendation suffer from serious problems of
generalization and oversmoothness, because of the underexplored negative
sampling method and the direct implanting of the off-the-shelf GNN models. In
this paper, we propose a succinct multi-network GNN-based neural model (NeMo)
for social recommendation. Compared with the existing methods, the proposed
model explores a generative negative sampling strategy, and leverages both the
positive and negative user-item interactions for users' interest propagation.
The experiments show that NeMo outperforms the state-of-the-art baselines on
various real-world benchmark datasets (e.g., by up to 38.8% in terms of
NDCG@15).
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 05:53:08 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Du",
"Boxin",
""
],
[
"Liu",
"Lihui",
""
],
[
"Xu",
"Jiejun",
""
],
[
"Wang",
"Fei",
""
],
[
"Tong",
"Hanghang",
""
]
] | TITLE: Neural Multi-network Diffusion towards Social Recommendation
ABSTRACT: Graph Neural Networks (GNNs) have been widely applied on a variety of
real-world applications, such as social recommendation. However, existing
GNN-based models on social recommendation suffer from serious problems of
generalization and oversmoothness, because of the underexplored negative
sampling method and the direct implanting of the off-the-shelf GNN models. In
this paper, we propose a succinct multi-network GNN-based neural model (NeMo)
for social recommendation. Compared with the existing methods, the proposed
model explores a generative negative sampling strategy, and leverages both the
positive and negative user-item interactions for users' interest propagation.
The experiments show that NeMo outperforms the state-of-the-art baselines on
various real-world benchmark datasets (e.g., by up to 38.8% in terms of
NDCG@15).
| no_new_dataset | 0.949716 |
2304.04997 | Sanghyun Kim | Sanghyun Kim, Deunsol Jung, Minsu Cho | Relational Context Learning for Human-Object Interaction Detection | accepted to CVPR2023 | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recent state-of-the-art methods for HOI detection typically build on
transformer architectures with two decoder branches, one for human-object pair
detection and the other for interaction classification. Such disentangled
transformers, however, may suffer from insufficient context exchange between
the branches and lead to a lack of context information for relational
reasoning, which is critical in discovering HOI instances. In this work, we
propose the multiplex relation network (MUREN) that performs rich context
exchange between three decoder branches using unary, pairwise, and ternary
relations of human, object, and interaction tokens. The proposed method learns
comprehensive relational contexts for discovering HOI instances, achieving
state-of-the-art performance on two standard benchmarks for HOI detection,
HICO-DET and V-COCO.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 06:01:10 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Kim",
"Sanghyun",
""
],
[
"Jung",
"Deunsol",
""
],
[
"Cho",
"Minsu",
""
]
] | TITLE: Relational Context Learning for Human-Object Interaction Detection
ABSTRACT: Recent state-of-the-art methods for HOI detection typically build on
transformer architectures with two decoder branches, one for human-object pair
detection and the other for interaction classification. Such disentangled
transformers, however, may suffer from insufficient context exchange between
the branches and lead to a lack of context information for relational
reasoning, which is critical in discovering HOI instances. In this work, we
propose the multiplex relation network (MUREN) that performs rich context
exchange between three decoder branches using unary, pairwise, and ternary
relations of human, object, and interaction tokens. The proposed method learns
comprehensive relational contexts for discovering HOI instances, achieving
state-of-the-art performance on two standard benchmarks for HOI detection,
HICO-DET and V-COCO.
| no_new_dataset | 0.950869 |
2304.05015 | Lanyun Zhu | Lanyun Zhu, Tianrun Chen, Jianxiong Yin, Simon See, Jun Liu | Continual Semantic Segmentation with Automatic Memory Sample Selection | Accepted to CVPR2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Continual Semantic Segmentation (CSS) extends static semantic segmentation by
incrementally introducing new classes for training. To alleviate the
catastrophic forgetting issue in CSS, a memory buffer that stores a small
number of samples from the previous classes is constructed for replay. However,
existing methods select the memory samples either randomly or based on a
single-factor-driven handcrafted strategy, which has no guarantee to be
optimal. In this work, we propose a novel memory sample selection mechanism
that selects informative samples for effective replay in a fully automatic way
by considering comprehensive factors including sample diversity and class
performance. Our mechanism regards the selection operation as a decision-making
process and learns an optimal selection policy that directly maximizes the
validation performance on a reward set. To facilitate the selection decision,
we design a novel state representation and a dual-stage action space. Our
extensive experiments on Pascal-VOC 2012 and ADE 20K datasets demonstrate the
effectiveness of our approach with state-of-the-art (SOTA) performance
achieved, outperforming the second-place one by 12.54% for the 6stage setting
on Pascal-VOC 2012.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 06:51:31 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Zhu",
"Lanyun",
""
],
[
"Chen",
"Tianrun",
""
],
[
"Yin",
"Jianxiong",
""
],
[
"See",
"Simon",
""
],
[
"Liu",
"Jun",
""
]
] | TITLE: Continual Semantic Segmentation with Automatic Memory Sample Selection
ABSTRACT: Continual Semantic Segmentation (CSS) extends static semantic segmentation by
incrementally introducing new classes for training. To alleviate the
catastrophic forgetting issue in CSS, a memory buffer that stores a small
number of samples from the previous classes is constructed for replay. However,
existing methods select the memory samples either randomly or based on a
single-factor-driven handcrafted strategy, which has no guarantee to be
optimal. In this work, we propose a novel memory sample selection mechanism
that selects informative samples for effective replay in a fully automatic way
by considering comprehensive factors including sample diversity and class
performance. Our mechanism regards the selection operation as a decision-making
process and learns an optimal selection policy that directly maximizes the
validation performance on a reward set. To facilitate the selection decision,
we design a novel state representation and a dual-stage action space. Our
extensive experiments on Pascal-VOC 2012 and ADE 20K datasets demonstrate the
effectiveness of our approach with state-of-the-art (SOTA) performance
achieved, outperforming the second-place one by 12.54% for the 6stage setting
on Pascal-VOC 2012.
| no_new_dataset | 0.949153 |
2304.05022 | Md Ishtyaq Mahmud | Md Ishtyaq Mahmud, Muntasir Mamun, Ahmed Abdelgawad | A Deep Analysis of Transfer Learning Based Breast Cancer Detection Using
Histopathology Images | 7 pages, Accepted by 10th International Conference on Signal
Processing and Integrated Networks (SPIN 2023) | null | null | null | eess.IV cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Breast cancer is one of the most common and dangerous cancers in women, while
it can also afflict men. Breast cancer treatment and detection are greatly
aided by the use of histopathological images since they contain sufficient
phenotypic data. A Deep Neural Network (DNN) is commonly employed to improve
accuracy and breast cancer detection. In our research, we have analyzed
pre-trained deep transfer learning models such as ResNet50, ResNet101, VGG16,
and VGG19 for detecting breast cancer using the 2453 histopathology images
dataset. Images in the dataset were separated into two categories: those with
invasive ductal carcinoma (IDC) and those without IDC. After analyzing the
transfer learning model, we found that ResNet50 outperformed other models,
achieving accuracy rates of 90.2%, Area under Curve (AUC) rates of 90.0%,
recall rates of 94.7%, and a marginal loss of 3.5%.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 07:17:55 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Mahmud",
"Md Ishtyaq",
""
],
[
"Mamun",
"Muntasir",
""
],
[
"Abdelgawad",
"Ahmed",
""
]
] | TITLE: A Deep Analysis of Transfer Learning Based Breast Cancer Detection Using
Histopathology Images
ABSTRACT: Breast cancer is one of the most common and dangerous cancers in women, while
it can also afflict men. Breast cancer treatment and detection are greatly
aided by the use of histopathological images since they contain sufficient
phenotypic data. A Deep Neural Network (DNN) is commonly employed to improve
accuracy and breast cancer detection. In our research, we have analyzed
pre-trained deep transfer learning models such as ResNet50, ResNet101, VGG16,
and VGG19 for detecting breast cancer using the 2453 histopathology images
dataset. Images in the dataset were separated into two categories: those with
invasive ductal carcinoma (IDC) and those without IDC. After analyzing the
transfer learning model, we found that ResNet50 outperformed other models,
achieving accuracy rates of 90.2%, Area under Curve (AUC) rates of 90.0%,
recall rates of 94.7%, and a marginal loss of 3.5%.
| no_new_dataset | 0.9462 |
2304.05041 | Matiss Rikters | Maija K\=ale and Mat\=iss Rikters | What Food Do We Tweet about on a Rainy Day? | null | Published in the proceedings of The 29th Annual Conference of the
Association for Natural Language Processing (NLP2023) | null | null | cs.SI cs.AI cs.CL | http://creativecommons.org/licenses/by/4.0/ | Food choice is a complex phenomenon shaped by factors such as taste,
ambience, culture or weather. In this paper, we explore food-related tweeting
in different weather conditions. We inspect a Latvian food tweet dataset
spanning the past decade in conjunction with a weather observation dataset
consisting of average temperature, precipitation, and other phenomena. We find
which weather conditions lead to specific food information sharing;
automatically classify tweet sentiment and discuss how it changes depending on
the weather. This research contributes to the growing area of large-scale
social network data understanding of food consumers' choices and perceptions.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 07:57:10 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Kāle",
"Maija",
""
],
[
"Rikters",
"Matīss",
""
]
] | TITLE: What Food Do We Tweet about on a Rainy Day?
ABSTRACT: Food choice is a complex phenomenon shaped by factors such as taste,
ambience, culture or weather. In this paper, we explore food-related tweeting
in different weather conditions. We inspect a Latvian food tweet dataset
spanning the past decade in conjunction with a weather observation dataset
consisting of average temperature, precipitation, and other phenomena. We find
which weather conditions lead to specific food information sharing;
automatically classify tweet sentiment and discuss how it changes depending on
the weather. This research contributes to the growing area of large-scale
social network data understanding of food consumers' choices and perceptions.
| no_new_dataset | 0.918845 |
2304.05051 | Yunpeng Han | Yunpeng Han, Lisai Zhang, Qingcai Chen, Zhijian Chen, Zhonghua Li,
Jianxin Yang, Zhao Cao | FashionSAP: Symbols and Attributes Prompt for Fine-grained Fashion
Vision-Language Pre-training | null | null | null | null | cs.CV cs.CL | http://creativecommons.org/licenses/by/4.0/ | Fashion vision-language pre-training models have shown efficacy for a wide
range of downstream tasks. However, general vision-language pre-training models
pay less attention to fine-grained domain features, while these features are
important in distinguishing the specific domain tasks from general tasks. We
propose a method for fine-grained fashion vision-language pre-training based on
fashion Symbols and Attributes Prompt (FashionSAP) to model fine-grained
multi-modalities fashion attributes and characteristics. Firstly, we propose
the fashion symbols, a novel abstract fashion concept layer, to represent
different fashion items and to generalize various kinds of fine-grained fashion
features, making modelling fine-grained attributes more effective. Secondly,
the attributes prompt method is proposed to make the model learn specific
attributes of fashion items explicitly. We design proper prompt templates
according to the format of fashion data. Comprehensive experiments are
conducted on two public fashion benchmarks, i.e., FashionGen and FashionIQ, and
FashionSAP gets SOTA performances for four popular fashion tasks. The ablation
study also shows the proposed abstract fashion symbols, and the attribute
prompt method enables the model to acquire fine-grained semantics in the
fashion domain effectively. The obvious performance gains from FashionSAP
provide a new baseline for future fashion task research.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 08:20:17 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Han",
"Yunpeng",
""
],
[
"Zhang",
"Lisai",
""
],
[
"Chen",
"Qingcai",
""
],
[
"Chen",
"Zhijian",
""
],
[
"Li",
"Zhonghua",
""
],
[
"Yang",
"Jianxin",
""
],
[
"Cao",
"Zhao",
""
]
] | TITLE: FashionSAP: Symbols and Attributes Prompt for Fine-grained Fashion
Vision-Language Pre-training
ABSTRACT: Fashion vision-language pre-training models have shown efficacy for a wide
range of downstream tasks. However, general vision-language pre-training models
pay less attention to fine-grained domain features, while these features are
important in distinguishing the specific domain tasks from general tasks. We
propose a method for fine-grained fashion vision-language pre-training based on
fashion Symbols and Attributes Prompt (FashionSAP) to model fine-grained
multi-modalities fashion attributes and characteristics. Firstly, we propose
the fashion symbols, a novel abstract fashion concept layer, to represent
different fashion items and to generalize various kinds of fine-grained fashion
features, making modelling fine-grained attributes more effective. Secondly,
the attributes prompt method is proposed to make the model learn specific
attributes of fashion items explicitly. We design proper prompt templates
according to the format of fashion data. Comprehensive experiments are
conducted on two public fashion benchmarks, i.e., FashionGen and FashionIQ, and
FashionSAP gets SOTA performances for four popular fashion tasks. The ablation
study also shows the proposed abstract fashion symbols, and the attribute
prompt method enables the model to acquire fine-grained semantics in the
fashion domain effectively. The obvious performance gains from FashionSAP
provide a new baseline for future fashion task research.
| no_new_dataset | 0.957118 |
2304.05064 | Lukas Holik | Tom\'a\v{s} Fiedor, Luk\'a\v{s} Hol\'ik, Martin Hru\v{s}ka, Adam
Rogalewicz, Juraj S\'i\v{c}, Pavol Vargov\v{c}\'ik | Reasoning about Regular Properties: A Comparative Study | null | null | null | null | cs.FL | http://creativecommons.org/licenses/by/4.0/ | Several new algorithms for deciding emptiness of Boolean combinations of
regular languages and of languages of alternating automata (AFA) have been
proposed recently, especially in the context of analysing regular expressions
and in string constraint solving. The new algorithms demonstrated a significant
potential, but they have never been systematically compared, neither among each
other nor with the state-of-the art implementations of existing
(non)deterministic automata-based methods. In this paper, we provide the first
such comparison as well as an overview of the existing algorithms and their
implementations. We collect a diverse benchmark mostly originating in or
related to practical problems from string constraint solving, analysing LTL
properties, and regular model checking, and evaluate collected implementations
on it. The results reveal the best tools and hint on what the best algorithms
and implementation techniques are. Roughly, although some advanced algorithms
are fast, such as antichain algorithms and reductions to IC3/PDR, they are not
as overwhelmingly dominant as sometimes presented and there is no clear winner.
The simplest NFA-based technology may be actually the best choice, depending on
the problem source and implementation style. Our findings should be highly
relevant for development of these techniques as well as for related fields such
as string constraint solving.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 08:54:22 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Fiedor",
"Tomáš",
""
],
[
"Holík",
"Lukáš",
""
],
[
"Hruška",
"Martin",
""
],
[
"Rogalewicz",
"Adam",
""
],
[
"Síč",
"Juraj",
""
],
[
"Vargovčík",
"Pavol",
""
]
] | TITLE: Reasoning about Regular Properties: A Comparative Study
ABSTRACT: Several new algorithms for deciding emptiness of Boolean combinations of
regular languages and of languages of alternating automata (AFA) have been
proposed recently, especially in the context of analysing regular expressions
and in string constraint solving. The new algorithms demonstrated a significant
potential, but they have never been systematically compared, neither among each
other nor with the state-of-the art implementations of existing
(non)deterministic automata-based methods. In this paper, we provide the first
such comparison as well as an overview of the existing algorithms and their
implementations. We collect a diverse benchmark mostly originating in or
related to practical problems from string constraint solving, analysing LTL
properties, and regular model checking, and evaluate collected implementations
on it. The results reveal the best tools and hint on what the best algorithms
and implementation techniques are. Roughly, although some advanced algorithms
are fast, such as antichain algorithms and reductions to IC3/PDR, they are not
as overwhelmingly dominant as sometimes presented and there is no clear winner.
The simplest NFA-based technology may be actually the best choice, depending on
the problem source and implementation style. Our findings should be highly
relevant for development of these techniques as well as for related fields such
as string constraint solving.
| no_new_dataset | 0.93646 |
2304.05080 | Sebastian Hafner | Sebastian Hafner, Yifang Ban, Andrea Nascetti | Investigating Imbalances Between SAR and Optical Utilization for
Multi-Modal Urban Mapping | 4 pages, 3 figures, accepted for publication in the JURSE 2023
Proceedings | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | Accurate urban maps provide essential information to support sustainable
urban development. Recent urban mapping methods use multi-modal deep neural
networks to fuse Synthetic Aperture Radar (SAR) and optical data. However,
multi-modal networks may rely on just one modality due to the greedy nature of
learning. In turn, the imbalanced utilization of modalities can negatively
affect the generalization ability of a network. In this paper, we investigate
the utilization of SAR and optical data for urban mapping. To that end, a
dual-branch network architecture using intermediate fusion modules to share
information between the uni-modal branches is utilized. A cut-off mechanism in
the fusion modules enables the stopping of information flow between the
branches, which is used to estimate the network's dependence on SAR and optical
data. While our experiments on the SEN12 Global Urban Mapping dataset show that
good performance can be achieved with conventional SAR-optical data fusion (F1
score = 0.682 $\pm$ 0.014), we also observed a clear under-utilization of
optical data. Therefore, future work is required to investigate whether a more
balanced utilization of SAR and optical data can lead to performance
improvements.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 09:22:51 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Hafner",
"Sebastian",
""
],
[
"Ban",
"Yifang",
""
],
[
"Nascetti",
"Andrea",
""
]
] | TITLE: Investigating Imbalances Between SAR and Optical Utilization for
Multi-Modal Urban Mapping
ABSTRACT: Accurate urban maps provide essential information to support sustainable
urban development. Recent urban mapping methods use multi-modal deep neural
networks to fuse Synthetic Aperture Radar (SAR) and optical data. However,
multi-modal networks may rely on just one modality due to the greedy nature of
learning. In turn, the imbalanced utilization of modalities can negatively
affect the generalization ability of a network. In this paper, we investigate
the utilization of SAR and optical data for urban mapping. To that end, a
dual-branch network architecture using intermediate fusion modules to share
information between the uni-modal branches is utilized. A cut-off mechanism in
the fusion modules enables the stopping of information flow between the
branches, which is used to estimate the network's dependence on SAR and optical
data. While our experiments on the SEN12 Global Urban Mapping dataset show that
good performance can be achieved with conventional SAR-optical data fusion (F1
score = 0.682 $\pm$ 0.014), we also observed a clear under-utilization of
optical data. Therefore, future work is required to investigate whether a more
balanced utilization of SAR and optical data can lead to performance
improvements.
| no_new_dataset | 0.945851 |
2304.05084 | Xin Chen | Xin Chen, Yuwen Qin, Weidong Zhao, Qiming Yang, Ningbo Cai, Kai Wu | A Self-attention Knowledge Domain Adaptation Network for Commercial
Lithium-ion Batteries State-of-health Estimation under Shallow Cycles | null | null | null | null | cs.LG eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate state-of-health (SOH) estimation is critical to guarantee the
safety, efficiency and reliability of battery-powered applications. Most SOH
estimation methods focus on the 0-100\% full state-of-charge (SOC) range that
has similar distributions. However, the batteries in real-world applications
usually work in the partial SOC range under shallow-cycle conditions and follow
different degradation profiles with no labeled data available, thus making SOH
estimation challenging. To estimate shallow-cycle battery SOH, a novel
unsupervised deep transfer learning method is proposed to bridge different
domains using self-attention distillation module and multi-kernel maximum mean
discrepancy technique. The proposed method automatically extracts
domain-variant features from charge curves to transfer knowledge from the
large-scale labeled full cycles to the unlabeled shallow cycles. The CALCE and
SNL battery datasets are employed to verify the effectiveness of the proposed
method to estimate the battery SOH for different SOC ranges, temperatures, and
discharge rates. The proposed method achieves a root-mean-square error within
2\% and outperforms other transfer learning methods for different SOC ranges.
When applied to batteries with different operating conditions and from
different manufacturers, the proposed method still exhibits superior SOH
estimation performance. The proposed method is the first attempt at accurately
estimating battery SOH under shallow-cycle conditions without needing a
full-cycle characteristic test.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 09:28:48 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Chen",
"Xin",
""
],
[
"Qin",
"Yuwen",
""
],
[
"Zhao",
"Weidong",
""
],
[
"Yang",
"Qiming",
""
],
[
"Cai",
"Ningbo",
""
],
[
"Wu",
"Kai",
""
]
] | TITLE: A Self-attention Knowledge Domain Adaptation Network for Commercial
Lithium-ion Batteries State-of-health Estimation under Shallow Cycles
ABSTRACT: Accurate state-of-health (SOH) estimation is critical to guarantee the
safety, efficiency and reliability of battery-powered applications. Most SOH
estimation methods focus on the 0-100\% full state-of-charge (SOC) range that
has similar distributions. However, the batteries in real-world applications
usually work in the partial SOC range under shallow-cycle conditions and follow
different degradation profiles with no labeled data available, thus making SOH
estimation challenging. To estimate shallow-cycle battery SOH, a novel
unsupervised deep transfer learning method is proposed to bridge different
domains using self-attention distillation module and multi-kernel maximum mean
discrepancy technique. The proposed method automatically extracts
domain-variant features from charge curves to transfer knowledge from the
large-scale labeled full cycles to the unlabeled shallow cycles. The CALCE and
SNL battery datasets are employed to verify the effectiveness of the proposed
method to estimate the battery SOH for different SOC ranges, temperatures, and
discharge rates. The proposed method achieves a root-mean-square error within
2\% and outperforms other transfer learning methods for different SOC ranges.
When applied to batteries with different operating conditions and from
different manufacturers, the proposed method still exhibits superior SOH
estimation performance. The proposed method is the first attempt at accurately
estimating battery SOH under shallow-cycle conditions without needing a
full-cycle characteristic test.
| no_new_dataset | 0.945951 |
2304.05090 | Luca Ciampi | Pawe{\l} Foszner, Agnieszka Szcz\k{e}sna, Luca Ciampi, Nicola Messina,
Adam Cygan, Bartosz Bizo\'n, Micha{\l} Cogiel, Dominik Golba, El\.zbieta
Macioszek, Micha{\l} Staniszewski | CrowdSim2: an Open Synthetic Benchmark for Object Detectors | Proceedings of the 18th International Joint Conference on Computer
Vision, Imaging and Computer Graphics Theory and Applications, 2023 | null | 10.5220/0011692500003417 | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Data scarcity has become one of the main obstacles to developing supervised
models based on Artificial Intelligence in Computer Vision. Indeed, Deep
Learning-based models systematically struggle when applied in new scenarios
never seen during training and may not be adequately tested in non-ordinary yet
crucial real-world situations. This paper presents and publicly releases
CrowdSim2, a new synthetic collection of images suitable for people and vehicle
detection gathered from a simulator based on the Unity graphical engine. It
consists of thousands of images gathered from various synthetic scenarios
resembling the real world, where we varied some factors of interest, such as
the weather conditions and the number of objects in the scenes. The labels are
automatically collected and consist of bounding boxes that precisely localize
objects belonging to the two object classes, leaving out humans from the
annotation pipeline. We exploited this new benchmark as a testing ground for
some state-of-the-art detectors, showing that our simulated scenarios can be a
valuable tool for measuring their performances in a controlled environment.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 09:35:57 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Foszner",
"Paweł",
""
],
[
"Szczęsna",
"Agnieszka",
""
],
[
"Ciampi",
"Luca",
""
],
[
"Messina",
"Nicola",
""
],
[
"Cygan",
"Adam",
""
],
[
"Bizoń",
"Bartosz",
""
],
[
"Cogiel",
"Michał",
""
],
[
"Golba",
"Dominik",
""
],
[
"Macioszek",
"Elżbieta",
""
],
[
"Staniszewski",
"Michał",
""
]
] | TITLE: CrowdSim2: an Open Synthetic Benchmark for Object Detectors
ABSTRACT: Data scarcity has become one of the main obstacles to developing supervised
models based on Artificial Intelligence in Computer Vision. Indeed, Deep
Learning-based models systematically struggle when applied in new scenarios
never seen during training and may not be adequately tested in non-ordinary yet
crucial real-world situations. This paper presents and publicly releases
CrowdSim2, a new synthetic collection of images suitable for people and vehicle
detection gathered from a simulator based on the Unity graphical engine. It
consists of thousands of images gathered from various synthetic scenarios
resembling the real world, where we varied some factors of interest, such as
the weather conditions and the number of objects in the scenes. The labels are
automatically collected and consist of bounding boxes that precisely localize
objects belonging to the two object classes, leaving out humans from the
annotation pipeline. We exploited this new benchmark as a testing ground for
some state-of-the-art detectors, showing that our simulated scenarios can be a
valuable tool for measuring their performances in a controlled environment.
| new_dataset | 0.931774 |
2304.05091 | Harry Jake Cunningham | Harry Jake Cunningham, Daniel Augusto de Souza, So Takao, Mark van der
Wilk, Marc Peter Deisenroth | Actually Sparse Variational Gaussian Processes | 14 pages, 5 figures, published in AISTATS 2023 | null | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by/4.0/ | Gaussian processes (GPs) are typically criticised for their unfavourable
scaling in both computational and memory requirements. For large datasets,
sparse GPs reduce these demands by conditioning on a small set of inducing
variables designed to summarise the data. In practice however, for large
datasets requiring many inducing variables, such as low-lengthscale spatial
data, even sparse GPs can become computationally expensive, limited by the
number of inducing variables one can use. In this work, we propose a new class
of inter-domain variational GP, constructed by projecting a GP onto a set of
compactly supported B-spline basis functions. The key benefit of our approach
is that the compact support of the B-spline basis functions admits the use of
sparse linear algebra to significantly speed up matrix operations and
drastically reduce the memory footprint. This allows us to very efficiently
model fast-varying spatial phenomena with tens of thousands of inducing
variables, where previous approaches failed.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 09:38:58 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Cunningham",
"Harry Jake",
""
],
[
"de Souza",
"Daniel Augusto",
""
],
[
"Takao",
"So",
""
],
[
"van der Wilk",
"Mark",
""
],
[
"Deisenroth",
"Marc Peter",
""
]
] | TITLE: Actually Sparse Variational Gaussian Processes
ABSTRACT: Gaussian processes (GPs) are typically criticised for their unfavourable
scaling in both computational and memory requirements. For large datasets,
sparse GPs reduce these demands by conditioning on a small set of inducing
variables designed to summarise the data. In practice however, for large
datasets requiring many inducing variables, such as low-lengthscale spatial
data, even sparse GPs can become computationally expensive, limited by the
number of inducing variables one can use. In this work, we propose a new class
of inter-domain variational GP, constructed by projecting a GP onto a set of
compactly supported B-spline basis functions. The key benefit of our approach
is that the compact support of the B-spline basis functions admits the use of
sparse linear algebra to significantly speed up matrix operations and
drastically reduce the memory footprint. This allows us to very efficiently
model fast-varying spatial phenomena with tens of thousands of inducing
variables, where previous approaches failed.
| no_new_dataset | 0.947235 |
2304.05096 | Hieu Le | Jingyi Xu, Hieu Le, Dimitris Samaras | Generating Features with Increased Crop-related Diversity for Few-Shot
Object Detection | Accepted to CVPR 23 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Two-stage object detectors generate object proposals and classify them to
detect objects in images. These proposals often do not contain the objects
perfectly but overlap with them in many possible ways, exhibiting great
variability in the difficulty levels of the proposals. Training a robust
classifier against this crop-related variability requires abundant training
data, which is not available in few-shot settings. To mitigate this issue, we
propose a novel variational autoencoder (VAE) based data generation model,
which is capable of generating data with increased crop-related diversity. The
main idea is to transform the latent space such latent codes with different
norms represent different crop-related variations. This allows us to generate
features with increased crop-related diversity in difficulty levels by simply
varying the latent norm. In particular, each latent code is rescaled such that
its norm linearly correlates with the IoU score of the input crop w.r.t. the
ground-truth box. Here the IoU score is a proxy that represents the difficulty
level of the crop. We train this VAE model on base classes conditioned on the
semantic code of each class and then use the trained model to generate features
for novel classes. In our experiments our generated features consistently
improve state-of-the-art few-shot object detection methods on the PASCAL VOC
and MS COCO datasets.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 09:47:21 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Xu",
"Jingyi",
""
],
[
"Le",
"Hieu",
""
],
[
"Samaras",
"Dimitris",
""
]
] | TITLE: Generating Features with Increased Crop-related Diversity for Few-Shot
Object Detection
ABSTRACT: Two-stage object detectors generate object proposals and classify them to
detect objects in images. These proposals often do not contain the objects
perfectly but overlap with them in many possible ways, exhibiting great
variability in the difficulty levels of the proposals. Training a robust
classifier against this crop-related variability requires abundant training
data, which is not available in few-shot settings. To mitigate this issue, we
propose a novel variational autoencoder (VAE) based data generation model,
which is capable of generating data with increased crop-related diversity. The
main idea is to transform the latent space such latent codes with different
norms represent different crop-related variations. This allows us to generate
features with increased crop-related diversity in difficulty levels by simply
varying the latent norm. In particular, each latent code is rescaled such that
its norm linearly correlates with the IoU score of the input crop w.r.t. the
ground-truth box. Here the IoU score is a proxy that represents the difficulty
level of the crop. We train this VAE model on base classes conditioned on the
semantic code of each class and then use the trained model to generate features
for novel classes. In our experiments our generated features consistently
improve state-of-the-art few-shot object detection methods on the PASCAL VOC
and MS COCO datasets.
| no_new_dataset | 0.951051 |
2304.05098 | Tianyuan Zhang | Tianyuan Zhang, Yisong Xiao, Xiaoya Zhang, Hao Li, Lu Wang | Benchmarking the Physical-world Adversarial Robustness of Vehicle
Detection | CVPR 2023 workshop | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adversarial attacks in the physical world can harm the robustness of
detection models. Evaluating the robustness of detection models in the physical
world can be challenging due to the time-consuming and labor-intensive nature
of many experiments. Thus, virtual simulation experiments can provide a
solution to this challenge. However, there is no unified detection benchmark
based on virtual simulation environment. To address this challenge, we proposed
an instant-level data generation pipeline based on the CARLA simulator. Using
this pipeline, we generated the DCI dataset and conducted extensive experiments
on three detection models and three physical adversarial attacks. The dataset
covers 7 continuous and 1 discrete scenes, with over 40 angles, 20 distances,
and 20,000 positions. The results indicate that Yolo v6 had strongest
resistance, with only a 6.59% average AP drop, and ASA was the most effective
attack algorithm with a 14.51% average AP reduction, twice that of other
algorithms. Static scenes had higher recognition AP, and results under
different weather conditions were similar. Adversarial attack algorithm
improvement may be approaching its 'limitation'.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 09:48:25 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Zhang",
"Tianyuan",
""
],
[
"Xiao",
"Yisong",
""
],
[
"Zhang",
"Xiaoya",
""
],
[
"Li",
"Hao",
""
],
[
"Wang",
"Lu",
""
]
] | TITLE: Benchmarking the Physical-world Adversarial Robustness of Vehicle
Detection
ABSTRACT: Adversarial attacks in the physical world can harm the robustness of
detection models. Evaluating the robustness of detection models in the physical
world can be challenging due to the time-consuming and labor-intensive nature
of many experiments. Thus, virtual simulation experiments can provide a
solution to this challenge. However, there is no unified detection benchmark
based on virtual simulation environment. To address this challenge, we proposed
an instant-level data generation pipeline based on the CARLA simulator. Using
this pipeline, we generated the DCI dataset and conducted extensive experiments
on three detection models and three physical adversarial attacks. The dataset
covers 7 continuous and 1 discrete scenes, with over 40 angles, 20 distances,
and 20,000 positions. The results indicate that Yolo v6 had strongest
resistance, with only a 6.59% average AP drop, and ASA was the most effective
attack algorithm with a 14.51% average AP reduction, twice that of other
algorithms. Static scenes had higher recognition AP, and results under
different weather conditions were similar. Adversarial attack algorithm
improvement may be approaching its 'limitation'.
| new_dataset | 0.954435 |
2304.05112 | Zhiwei Yang | Zhiwei Yang, Jing Liu, Zhaoyang Wu, Peng Wu, Xiaotao Liu | Video Event Restoration Based on Keyframes for Video Anomaly Detection | Accepted by CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Video anomaly detection (VAD) is a significant computer vision problem.
Existing deep neural network (DNN) based VAD methods mostly follow the route of
frame reconstruction or frame prediction. However, the lack of mining and
learning of higher-level visual features and temporal context relationships in
videos limits the further performance of these two approaches. Inspired by
video codec theory, we introduce a brand-new VAD paradigm to break through
these limitations: First, we propose a new task of video event restoration
based on keyframes. Encouraging DNN to infer missing multiple frames based on
video keyframes so as to restore a video event, which can more effectively
motivate DNN to mine and learn potential higher-level visual features and
comprehensive temporal context relationships in the video. To this end, we
propose a novel U-shaped Swin Transformer Network with Dual Skip Connections
(USTN-DSC) for video event restoration, where a cross-attention and a temporal
upsampling residual skip connection are introduced to further assist in
restoring complex static and dynamic motion object features in the video. In
addition, we propose a simple and effective adjacent frame difference loss to
constrain the motion consistency of the video sequence. Extensive experiments
on benchmarks demonstrate that USTN-DSC outperforms most existing methods,
validating the effectiveness of our method.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 10:13:19 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Yang",
"Zhiwei",
""
],
[
"Liu",
"Jing",
""
],
[
"Wu",
"Zhaoyang",
""
],
[
"Wu",
"Peng",
""
],
[
"Liu",
"Xiaotao",
""
]
] | TITLE: Video Event Restoration Based on Keyframes for Video Anomaly Detection
ABSTRACT: Video anomaly detection (VAD) is a significant computer vision problem.
Existing deep neural network (DNN) based VAD methods mostly follow the route of
frame reconstruction or frame prediction. However, the lack of mining and
learning of higher-level visual features and temporal context relationships in
videos limits the further performance of these two approaches. Inspired by
video codec theory, we introduce a brand-new VAD paradigm to break through
these limitations: First, we propose a new task of video event restoration
based on keyframes. Encouraging DNN to infer missing multiple frames based on
video keyframes so as to restore a video event, which can more effectively
motivate DNN to mine and learn potential higher-level visual features and
comprehensive temporal context relationships in the video. To this end, we
propose a novel U-shaped Swin Transformer Network with Dual Skip Connections
(USTN-DSC) for video event restoration, where a cross-attention and a temporal
upsampling residual skip connection are introduced to further assist in
restoring complex static and dynamic motion object features in the video. In
addition, we propose a simple and effective adjacent frame difference loss to
constrain the motion consistency of the video sequence. Extensive experiments
on benchmarks demonstrate that USTN-DSC outperforms most existing methods,
validating the effectiveness of our method.
| no_new_dataset | 0.944331 |
2304.05115 | Jianfei Zhang | Jianfei Zhang and Mathieu Rosenbaum | Towards systematic intraday news screening: a liquidity-focused approach | null | null | null | null | q-fin.TR cs.LG q-fin.CP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | News can convey bearish or bullish views on financial assets. Institutional
investors need to evaluate automatically the implied news sentiment based on
textual data. Given the huge amount of news articles published each day, most
of which are neutral, we present a systematic news screening method to identify
the ``true'' impactful ones, aiming for more effective development of news
sentiment learning methods. Based on several liquidity-driven variables,
including volatility, turnover, bid-ask spread, and book size, we associate
each 5-min time bin to one of two specific liquidity modes. One represents the
``calm'' state at which the market stays for most of the time and the other,
featured with relatively higher levels of volatility and trading volume,
describes the regime driven by some exogenous events. Then we focus on the
moments where the liquidity mode switches from the former to the latter and
consider the news articles published nearby impactful. We apply naive Bayes on
these filtered samples for news sentiment classification as an illustrative
example. We show that the screened dataset leads to more effective feature
capturing and thus superior performance on short-term asset return prediction
compared to the original dataset.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 10:14:48 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Zhang",
"Jianfei",
""
],
[
"Rosenbaum",
"Mathieu",
""
]
] | TITLE: Towards systematic intraday news screening: a liquidity-focused approach
ABSTRACT: News can convey bearish or bullish views on financial assets. Institutional
investors need to evaluate automatically the implied news sentiment based on
textual data. Given the huge amount of news articles published each day, most
of which are neutral, we present a systematic news screening method to identify
the ``true'' impactful ones, aiming for more effective development of news
sentiment learning methods. Based on several liquidity-driven variables,
including volatility, turnover, bid-ask spread, and book size, we associate
each 5-min time bin to one of two specific liquidity modes. One represents the
``calm'' state at which the market stays for most of the time and the other,
featured with relatively higher levels of volatility and trading volume,
describes the regime driven by some exogenous events. Then we focus on the
moments where the liquidity mode switches from the former to the latter and
consider the news articles published nearby impactful. We apply naive Bayes on
these filtered samples for news sentiment classification as an illustrative
example. We show that the screened dataset leads to more effective feature
capturing and thus superior performance on short-term asset return prediction
compared to the original dataset.
| no_new_dataset | 0.952794 |
2304.05118 | Ewelina Rupnik | Ewelina Rupnik and Marc Pierrot-Deseilligny | Pointless Global Bundle Adjustment With Relative Motions Hessians | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Bundle adjustment (BA) is the standard way to optimise camera poses and to
produce sparse representations of a scene. However, as the number of camera
poses and features grows, refinement through bundle adjustment becomes
inefficient. Inspired by global motion averaging methods, we propose a new
bundle adjustment objective which does not rely on image features' reprojection
errors yet maintains precision on par with classical BA. Our method averages
over relative motions while implicitly incorporating the contribution of the
structure in the adjustment. To that end, we weight the objective function by
local hessian matrices - a by-product of local bundle adjustments performed on
relative motions (e.g., pairs or triplets) during the pose initialisation step.
Such hessians are extremely rich as they encapsulate both the features' random
errors and the geometric configuration between the cameras. These pieces of
information propagated to the global frame help to guide the final optimisation
in a more rigorous way. We argue that this approach is an upgraded version of
the motion averaging approach and demonstrate its effectiveness on both
photogrammetric datasets and computer vision benchmarks.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 10:20:32 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Rupnik",
"Ewelina",
""
],
[
"Pierrot-Deseilligny",
"Marc",
""
]
] | TITLE: Pointless Global Bundle Adjustment With Relative Motions Hessians
ABSTRACT: Bundle adjustment (BA) is the standard way to optimise camera poses and to
produce sparse representations of a scene. However, as the number of camera
poses and features grows, refinement through bundle adjustment becomes
inefficient. Inspired by global motion averaging methods, we propose a new
bundle adjustment objective which does not rely on image features' reprojection
errors yet maintains precision on par with classical BA. Our method averages
over relative motions while implicitly incorporating the contribution of the
structure in the adjustment. To that end, we weight the objective function by
local hessian matrices - a by-product of local bundle adjustments performed on
relative motions (e.g., pairs or triplets) during the pose initialisation step.
Such hessians are extremely rich as they encapsulate both the features' random
errors and the geometric configuration between the cameras. These pieces of
information propagated to the global frame help to guide the final optimisation
in a more rigorous way. We argue that this approach is an upgraded version of
the motion averaging approach and demonstrate its effectiveness on both
photogrammetric datasets and computer vision benchmarks.
| no_new_dataset | 0.946101 |
2304.05121 | Kisub Kim | Kisub Kim, Ferdian Thung, Ting Zhang, Ivana Clairine Irsan, Ratnadira
Widyasari, Zhou Yang, David Lo | APISENS- Sentiment Scoring Tool for APIs with Crowd-Knowledge | null | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | Utilizing pre-existing software artifacts, such as libraries and Application
Programming Interfaces (APIs), is crucial for software development efficiency.
However, the abundance of artifacts that provide similar functionality can lead
to confusion among developers, resulting in a challenge for proper selection
and implementation. Through our preliminary investigation, we found that
utilizing the collective knowledge of a crowd can greatly assist developers in
acquiring a thorough and complete understanding of the complexities involved in
the software development process. Especially as emotions are an inseparable
part of human nature, it influences developers' activities. In this regard, we
attempt to build a tool that can retrieve sentiment information for software
APIs so that developers can determine APIs to utilize for their tasks. We
employ the dataset from the most popular platforms (i.e., Twitter and YouTube)
to build our research prototype. The source code, tool, and demo video are
available on GitHub at \url{https://github.com/FalconLK/APISens}.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 10:26:12 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Kim",
"Kisub",
""
],
[
"Thung",
"Ferdian",
""
],
[
"Zhang",
"Ting",
""
],
[
"Irsan",
"Ivana Clairine",
""
],
[
"Widyasari",
"Ratnadira",
""
],
[
"Yang",
"Zhou",
""
],
[
"Lo",
"David",
""
]
] | TITLE: APISENS- Sentiment Scoring Tool for APIs with Crowd-Knowledge
ABSTRACT: Utilizing pre-existing software artifacts, such as libraries and Application
Programming Interfaces (APIs), is crucial for software development efficiency.
However, the abundance of artifacts that provide similar functionality can lead
to confusion among developers, resulting in a challenge for proper selection
and implementation. Through our preliminary investigation, we found that
utilizing the collective knowledge of a crowd can greatly assist developers in
acquiring a thorough and complete understanding of the complexities involved in
the software development process. Especially as emotions are an inseparable
part of human nature, it influences developers' activities. In this regard, we
attempt to build a tool that can retrieve sentiment information for software
APIs so that developers can determine APIs to utilize for their tasks. We
employ the dataset from the most popular platforms (i.e., Twitter and YouTube)
to build our research prototype. The source code, tool, and demo video are
available on GitHub at \url{https://github.com/FalconLK/APISens}.
| no_new_dataset | 0.9463 |
2304.05135 | Yue Cui | Yue Cui, Syed Irfan Ali Meerza, Zhuohang Li, Luyang Liu, Jiaxin Zhang,
Jian Liu | RecUP-FL: Reconciling Utility and Privacy in Federated Learning via
User-configurable Privacy Defense | null | null | 10.1145/3579856.3582819 | null | cs.LG cs.CR | http://creativecommons.org/licenses/by/4.0/ | Federated learning (FL) provides a variety of privacy advantages by allowing
clients to collaboratively train a model without sharing their private data.
However, recent studies have shown that private information can still be leaked
through shared gradients. To further minimize the risk of privacy leakage,
existing defenses usually require clients to locally modify their gradients
(e.g., differential privacy) prior to sharing with the server. While these
approaches are effective in certain cases, they regard the entire data as a
single entity to protect, which usually comes at a large cost in model utility.
In this paper, we seek to reconcile utility and privacy in FL by proposing a
user-configurable privacy defense, RecUP-FL, that can better focus on the
user-specified sensitive attributes while obtaining significant improvements in
utility over traditional defenses. Moreover, we observe that existing inference
attacks often rely on a machine learning model to extract the private
information (e.g., attributes). We thus formulate such a privacy defense as an
adversarial learning problem, where RecUP-FL generates slight perturbations
that can be added to the gradients before sharing to fool adversary models. To
improve the transferability to un-queryable black-box adversary models,
inspired by the idea of meta-learning, RecUP-FL forms a model zoo containing a
set of substitute models and iteratively alternates between simulations of the
white-box and the black-box adversarial attack scenarios to generate
perturbations. Extensive experiments on four datasets under various adversarial
settings (both attribute inference attack and data reconstruction attack) show
that RecUP-FL can meet user-specified privacy constraints over the sensitive
attributes while significantly improving the model utility compared with
state-of-the-art privacy defenses.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 10:59:45 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Cui",
"Yue",
""
],
[
"Meerza",
"Syed Irfan Ali",
""
],
[
"Li",
"Zhuohang",
""
],
[
"Liu",
"Luyang",
""
],
[
"Zhang",
"Jiaxin",
""
],
[
"Liu",
"Jian",
""
]
] | TITLE: RecUP-FL: Reconciling Utility and Privacy in Federated Learning via
User-configurable Privacy Defense
ABSTRACT: Federated learning (FL) provides a variety of privacy advantages by allowing
clients to collaboratively train a model without sharing their private data.
However, recent studies have shown that private information can still be leaked
through shared gradients. To further minimize the risk of privacy leakage,
existing defenses usually require clients to locally modify their gradients
(e.g., differential privacy) prior to sharing with the server. While these
approaches are effective in certain cases, they regard the entire data as a
single entity to protect, which usually comes at a large cost in model utility.
In this paper, we seek to reconcile utility and privacy in FL by proposing a
user-configurable privacy defense, RecUP-FL, that can better focus on the
user-specified sensitive attributes while obtaining significant improvements in
utility over traditional defenses. Moreover, we observe that existing inference
attacks often rely on a machine learning model to extract the private
information (e.g., attributes). We thus formulate such a privacy defense as an
adversarial learning problem, where RecUP-FL generates slight perturbations
that can be added to the gradients before sharing to fool adversary models. To
improve the transferability to un-queryable black-box adversary models,
inspired by the idea of meta-learning, RecUP-FL forms a model zoo containing a
set of substitute models and iteratively alternates between simulations of the
white-box and the black-box adversarial attack scenarios to generate
perturbations. Extensive experiments on four datasets under various adversarial
settings (both attribute inference attack and data reconstruction attack) show
that RecUP-FL can meet user-specified privacy constraints over the sensitive
attributes while significantly improving the model utility compared with
state-of-the-art privacy defenses.
| no_new_dataset | 0.943295 |
2304.05137 | Markus Buehler | Wei Lu, Nic A. Lee, Markus J. Buehler | Modeling and design of heterogeneous hierarchical bioinspired spider web
structures using generative deep learning and additive manufacturing | null | null | null | null | cs.LG cond-mat.soft nlin.AO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Spider webs are incredible biological structures, comprising thin but strong
silk filament and arranged into complex hierarchical architectures with
striking mechanical properties (e.g., lightweight but high strength, achieving
diverse mechanical responses). While simple 2D orb webs can easily be mimicked,
the modeling and synthesis of 3D-based web structures remain challenging,
partly due to the rich set of design features. Here we provide a detailed
analysis of the heterogenous graph structures of spider webs, and use deep
learning as a way to model and then synthesize artificial, bio-inspired 3D web
structures. The generative AI models are conditioned based on key geometric
parameters (including average edge length, number of nodes, average node
degree, and others). To identify graph construction principles, we use
inductive representation sampling of large experimentally determined spider web
graphs, to yield a dataset that is used to train three conditional generative
models: 1) An analog diffusion model inspired by nonequilibrium thermodynamics,
with sparse neighbor representation, 2) a discrete diffusion model with full
neighbor representation, and 3) an autoregressive transformer architecture with
full neighbor representation. All three models are scalable, produce complex,
de novo bio-inspired spider web mimics, and successfully construct graphs that
meet the design objectives. We further propose algorithm that assembles web
samples produced by the generative models into larger-scale structures based on
a series of geometric design targets, including helical and parametric shapes,
mimicking, and extending natural design principles towards integration with
diverging engineering objectives. Several webs are manufactured using 3D
printing and tested to assess mechanical properties.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 11:03:27 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Lu",
"Wei",
""
],
[
"Lee",
"Nic A.",
""
],
[
"Buehler",
"Markus J.",
""
]
] | TITLE: Modeling and design of heterogeneous hierarchical bioinspired spider web
structures using generative deep learning and additive manufacturing
ABSTRACT: Spider webs are incredible biological structures, comprising thin but strong
silk filament and arranged into complex hierarchical architectures with
striking mechanical properties (e.g., lightweight but high strength, achieving
diverse mechanical responses). While simple 2D orb webs can easily be mimicked,
the modeling and synthesis of 3D-based web structures remain challenging,
partly due to the rich set of design features. Here we provide a detailed
analysis of the heterogenous graph structures of spider webs, and use deep
learning as a way to model and then synthesize artificial, bio-inspired 3D web
structures. The generative AI models are conditioned based on key geometric
parameters (including average edge length, number of nodes, average node
degree, and others). To identify graph construction principles, we use
inductive representation sampling of large experimentally determined spider web
graphs, to yield a dataset that is used to train three conditional generative
models: 1) An analog diffusion model inspired by nonequilibrium thermodynamics,
with sparse neighbor representation, 2) a discrete diffusion model with full
neighbor representation, and 3) an autoregressive transformer architecture with
full neighbor representation. All three models are scalable, produce complex,
de novo bio-inspired spider web mimics, and successfully construct graphs that
meet the design objectives. We further propose algorithm that assembles web
samples produced by the generative models into larger-scale structures based on
a series of geometric design targets, including helical and parametric shapes,
mimicking, and extending natural design principles towards integration with
diverging engineering objectives. Several webs are manufactured using 3D
printing and tested to assess mechanical properties.
| no_new_dataset | 0.825941 |
2304.05139 | Dan Ruta | Dan Ruta, Andrew Gilbert, John Collomosse, Eli Shechtman, Nicholas
Kolkin | NeAT: Neural Artistic Tracing for Beautiful Style Transfer | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Style transfer is the task of reproducing the semantic contents of a source
image in the artistic style of a second target image. In this paper, we present
NeAT, a new state-of-the art feed-forward style transfer method. We
re-formulate feed-forward style transfer as image editing, rather than image
generation, resulting in a model which improves over the state-of-the-art in
both preserving the source content and matching the target style. An important
component of our model's success is identifying and fixing "style halos", a
commonly occurring artefact across many style transfer techniques. In addition
to training and testing on standard datasets, we introduce the BBST-4M dataset,
a new, large scale, high resolution dataset of 4M images. As a component of
curating this data, we present a novel model able to classify if an image is
stylistic. We use BBST-4M to improve and measure the generalization of NeAT
across a huge variety of styles. Not only does NeAT offer state-of-the-art
quality and generalization, it is designed and trained for fast inference at
high resolution.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 11:08:13 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Ruta",
"Dan",
""
],
[
"Gilbert",
"Andrew",
""
],
[
"Collomosse",
"John",
""
],
[
"Shechtman",
"Eli",
""
],
[
"Kolkin",
"Nicholas",
""
]
] | TITLE: NeAT: Neural Artistic Tracing for Beautiful Style Transfer
ABSTRACT: Style transfer is the task of reproducing the semantic contents of a source
image in the artistic style of a second target image. In this paper, we present
NeAT, a new state-of-the art feed-forward style transfer method. We
re-formulate feed-forward style transfer as image editing, rather than image
generation, resulting in a model which improves over the state-of-the-art in
both preserving the source content and matching the target style. An important
component of our model's success is identifying and fixing "style halos", a
commonly occurring artefact across many style transfer techniques. In addition
to training and testing on standard datasets, we introduce the BBST-4M dataset,
a new, large scale, high resolution dataset of 4M images. As a component of
curating this data, we present a novel model able to classify if an image is
stylistic. We use BBST-4M to improve and measure the generalization of NeAT
across a huge variety of styles. Not only does NeAT offer state-of-the-art
quality and generalization, it is designed and trained for fast inference at
high resolution.
| new_dataset | 0.958924 |
2304.05152 | Shiyu Tang | Shiyu Tang, Ting Sun, Juncai Peng, Guowei Chen, Yuying Hao, Manhui
Lin, Zhihong Xiao, Jiangbin You, Yi Liu | PP-MobileSeg: Explore the Fast and Accurate Semantic Segmentation Model
on Mobile Devices | 8 pages, 3 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The success of transformers in computer vision has led to several attempts to
adapt them for mobile devices, but their performance remains unsatisfactory in
some real-world applications. To address this issue, we propose PP-MobileSeg, a
semantic segmentation model that achieves state-of-the-art performance on
mobile devices. PP-MobileSeg comprises three novel parts: the StrideFormer
backbone, the Aggregated Attention Module (AAM), and the Valid Interpolate
Module (VIM). The four-stage StrideFormer backbone is built with MV3 blocks and
strided SEA attention, and it is able to extract rich semantic and detailed
features with minimal parameter overhead. The AAM first filters the detailed
features through semantic feature ensemble voting and then combines them with
semantic features to enhance the semantic information. Furthermore, we proposed
VIM to upsample the downsampled feature to the resolution of the input image.
It significantly reduces model latency by only interpolating classes present in
the final prediction, which is the most significant contributor to overall
model latency. Extensive experiments show that PP-MobileSeg achieves a superior
tradeoff between accuracy, model size, and latency compared to other methods.
On the ADE20K dataset, PP-MobileSeg achieves 1.57% higher accuracy in mIoU than
SeaFormer-Base with 32.9% fewer parameters and 42.3% faster acceleration on
Qualcomm Snapdragon 855. Source codes are available at
https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.8.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 11:43:10 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Tang",
"Shiyu",
""
],
[
"Sun",
"Ting",
""
],
[
"Peng",
"Juncai",
""
],
[
"Chen",
"Guowei",
""
],
[
"Hao",
"Yuying",
""
],
[
"Lin",
"Manhui",
""
],
[
"Xiao",
"Zhihong",
""
],
[
"You",
"Jiangbin",
""
],
[
"Liu",
"Yi",
""
]
] | TITLE: PP-MobileSeg: Explore the Fast and Accurate Semantic Segmentation Model
on Mobile Devices
ABSTRACT: The success of transformers in computer vision has led to several attempts to
adapt them for mobile devices, but their performance remains unsatisfactory in
some real-world applications. To address this issue, we propose PP-MobileSeg, a
semantic segmentation model that achieves state-of-the-art performance on
mobile devices. PP-MobileSeg comprises three novel parts: the StrideFormer
backbone, the Aggregated Attention Module (AAM), and the Valid Interpolate
Module (VIM). The four-stage StrideFormer backbone is built with MV3 blocks and
strided SEA attention, and it is able to extract rich semantic and detailed
features with minimal parameter overhead. The AAM first filters the detailed
features through semantic feature ensemble voting and then combines them with
semantic features to enhance the semantic information. Furthermore, we proposed
VIM to upsample the downsampled feature to the resolution of the input image.
It significantly reduces model latency by only interpolating classes present in
the final prediction, which is the most significant contributor to overall
model latency. Extensive experiments show that PP-MobileSeg achieves a superior
tradeoff between accuracy, model size, and latency compared to other methods.
On the ADE20K dataset, PP-MobileSeg achieves 1.57% higher accuracy in mIoU than
SeaFormer-Base with 32.9% fewer parameters and 42.3% faster acceleration on
Qualcomm Snapdragon 855. Source codes are available at
https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.8.
| no_new_dataset | 0.949669 |
2304.05165 | Mengyao Xie | Mengyao Xie, Zongbo Han, Changqing Zhang, Yichen Bai, Qinghua Hu | Exploring and Exploiting Uncertainty for Incomplete Multi-View
Classification | CVPR | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Classifying incomplete multi-view data is inevitable since arbitrary view
missing widely exists in real-world applications. Although great progress has
been achieved, existing incomplete multi-view methods are still difficult to
obtain a trustworthy prediction due to the relatively high uncertainty nature
of missing views. First, the missing view is of high uncertainty, and thus it
is not reasonable to provide a single deterministic imputation. Second, the
quality of the imputed data itself is of high uncertainty. To explore and
exploit the uncertainty, we propose an Uncertainty-induced Incomplete
Multi-View Data Classification (UIMC) model to classify the incomplete
multi-view data under a stable and reliable framework. We construct a
distribution and sample multiple times to characterize the uncertainty of
missing views, and adaptively utilize them according to the sampling quality.
Accordingly, the proposed method realizes more perceivable imputation and
controllable fusion. Specifically, we model each missing data with a
distribution conditioning on the available views and thus introducing
uncertainty. Then an evidence-based fusion strategy is employed to guarantee
the trustworthy integration of the imputed views. Extensive experiments are
conducted on multiple benchmark data sets and our method establishes a
state-of-the-art performance in terms of both performance and trustworthiness.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 11:57:48 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Xie",
"Mengyao",
""
],
[
"Han",
"Zongbo",
""
],
[
"Zhang",
"Changqing",
""
],
[
"Bai",
"Yichen",
""
],
[
"Hu",
"Qinghua",
""
]
] | TITLE: Exploring and Exploiting Uncertainty for Incomplete Multi-View
Classification
ABSTRACT: Classifying incomplete multi-view data is inevitable since arbitrary view
missing widely exists in real-world applications. Although great progress has
been achieved, existing incomplete multi-view methods are still difficult to
obtain a trustworthy prediction due to the relatively high uncertainty nature
of missing views. First, the missing view is of high uncertainty, and thus it
is not reasonable to provide a single deterministic imputation. Second, the
quality of the imputed data itself is of high uncertainty. To explore and
exploit the uncertainty, we propose an Uncertainty-induced Incomplete
Multi-View Data Classification (UIMC) model to classify the incomplete
multi-view data under a stable and reliable framework. We construct a
distribution and sample multiple times to characterize the uncertainty of
missing views, and adaptively utilize them according to the sampling quality.
Accordingly, the proposed method realizes more perceivable imputation and
controllable fusion. Specifically, we model each missing data with a
distribution conditioning on the available views and thus introducing
uncertainty. Then an evidence-based fusion strategy is employed to guarantee
the trustworthy integration of the imputed views. Extensive experiments are
conducted on multiple benchmark data sets and our method establishes a
state-of-the-art performance in terms of both performance and trustworthiness.
| no_new_dataset | 0.945851 |
2304.05173 | Ahmet Iscen | Ahmet Iscen, Alireza Fathi, Cordelia Schmid | Improving Image Recognition by Retrieving from Web-Scale Image-Text Data | Accepted to CVPR 2023 | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Retrieval augmented models are becoming increasingly popular for computer
vision tasks after their recent success in NLP problems. The goal is to enhance
the recognition capabilities of the model by retrieving similar examples for
the visual input from an external memory set. In this work, we introduce an
attention-based memory module, which learns the importance of each retrieved
example from the memory. Compared to existing approaches, our method removes
the influence of the irrelevant retrieved examples, and retains those that are
beneficial to the input query. We also thoroughly study various ways of
constructing the memory dataset. Our experiments show the benefit of using a
massive-scale memory dataset of 1B image-text pairs, and demonstrate the
performance of different memory representations. We evaluate our method in
three different classification tasks, namely long-tailed recognition, learning
with noisy labels, and fine-grained classification, and show that it achieves
state-of-the-art accuracies in ImageNet-LT, Places-LT and Webvision datasets.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 12:12:05 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Iscen",
"Ahmet",
""
],
[
"Fathi",
"Alireza",
""
],
[
"Schmid",
"Cordelia",
""
]
] | TITLE: Improving Image Recognition by Retrieving from Web-Scale Image-Text Data
ABSTRACT: Retrieval augmented models are becoming increasingly popular for computer
vision tasks after their recent success in NLP problems. The goal is to enhance
the recognition capabilities of the model by retrieving similar examples for
the visual input from an external memory set. In this work, we introduce an
attention-based memory module, which learns the importance of each retrieved
example from the memory. Compared to existing approaches, our method removes
the influence of the irrelevant retrieved examples, and retains those that are
beneficial to the input query. We also thoroughly study various ways of
constructing the memory dataset. Our experiments show the benefit of using a
massive-scale memory dataset of 1B image-text pairs, and demonstrate the
performance of different memory representations. We evaluate our method in
three different classification tasks, namely long-tailed recognition, learning
with noisy labels, and fine-grained classification, and show that it achieves
state-of-the-art accuracies in ImageNet-LT, Places-LT and Webvision datasets.
| no_new_dataset | 0.907722 |
2304.05176 | YanMing Hu | YanMing Hu, Chuan Chen, BoWen Deng, YuJing Lai, Hao Lin, ZiBin Zheng
and Jing Bian | Decoupling anomaly discrimination and representation learning:
self-supervised learning for anomaly detection on attributed graph | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Anomaly detection on attributed graphs is a crucial topic for its practical
application. Existing methods suffer from semantic mixture and imbalance issue
because they mainly focus on anomaly discrimination, ignoring representation
learning. It conflicts with the assortativity assumption that anomalous nodes
commonly connect with normal nodes directly. Additionally, there are far fewer
anomalous nodes than normal nodes, indicating a long-tailed data distribution.
To address these challenges, a unique algorithm,Decoupled Self-supervised
Learning forAnomalyDetection (DSLAD), is proposed in this paper. DSLAD is a
self-supervised method with anomaly discrimination and representation learning
decoupled for anomaly detection. DSLAD employs bilinear pooling and masked
autoencoder as the anomaly discriminators. By decoupling anomaly discrimination
and representation learning, a balanced feature space is constructed, in which
nodes are more semantically discriminative, as well as imbalance issue can be
resolved. Experiments conducted on various six benchmark datasets reveal the
effectiveness of DSLAD.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 12:23:40 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Hu",
"YanMing",
""
],
[
"Chen",
"Chuan",
""
],
[
"Deng",
"BoWen",
""
],
[
"Lai",
"YuJing",
""
],
[
"Lin",
"Hao",
""
],
[
"Zheng",
"ZiBin",
""
],
[
"Bian",
"Jing",
""
]
] | TITLE: Decoupling anomaly discrimination and representation learning:
self-supervised learning for anomaly detection on attributed graph
ABSTRACT: Anomaly detection on attributed graphs is a crucial topic for its practical
application. Existing methods suffer from semantic mixture and imbalance issue
because they mainly focus on anomaly discrimination, ignoring representation
learning. It conflicts with the assortativity assumption that anomalous nodes
commonly connect with normal nodes directly. Additionally, there are far fewer
anomalous nodes than normal nodes, indicating a long-tailed data distribution.
To address these challenges, a unique algorithm,Decoupled Self-supervised
Learning forAnomalyDetection (DSLAD), is proposed in this paper. DSLAD is a
self-supervised method with anomaly discrimination and representation learning
decoupled for anomaly detection. DSLAD employs bilinear pooling and masked
autoencoder as the anomaly discriminators. By decoupling anomaly discrimination
and representation learning, a balanced feature space is constructed, in which
nodes are more semantically discriminative, as well as imbalance issue can be
resolved. Experiments conducted on various six benchmark datasets reveal the
effectiveness of DSLAD.
| no_new_dataset | 0.949949 |
2304.05193 | Jian Wang Jornbowrl | Jian Wang, Shangqing Liu, Xiaofei Xie, Yi Li | Evaluating AIGC Detectors on Code Content | null | null | null | null | cs.SE | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Artificial Intelligence Generated Content (AIGC) has garnered considerable
attention for its impressive performance, with ChatGPT emerging as a leading
AIGC model that produces high-quality responses across various applications,
including software development and maintenance. Despite its potential, the
misuse of ChatGPT poses significant concerns, especially in education and
safetycritical domains. Numerous AIGC detectors have been developed and
evaluated on natural language data. However, their performance on code-related
content generated by ChatGPT remains unexplored. To fill this gap, in this
paper, we present the first empirical study on evaluating existing AIGC
detectors in the software domain. We created a comprehensive dataset including
492.5K samples comprising code-related content produced by ChatGPT,
encompassing popular software activities like Q&A (115K), code summarization
(126K), and code generation (226.5K). We evaluated six AIGC detectors,
including three commercial and three open-source solutions, assessing their
performance on this dataset. Additionally, we conducted a human study to
understand human detection capabilities and compare them with the existing AIGC
detectors. Our results indicate that AIGC detectors demonstrate lower
performance on code-related data compared to natural language data. Fine-tuning
can enhance detector performance, especially for content within the same
domain; but generalization remains a challenge. The human evaluation reveals
that detection by humans is quite challenging.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 12:54:42 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wang",
"Jian",
""
],
[
"Liu",
"Shangqing",
""
],
[
"Xie",
"Xiaofei",
""
],
[
"Li",
"Yi",
""
]
] | TITLE: Evaluating AIGC Detectors on Code Content
ABSTRACT: Artificial Intelligence Generated Content (AIGC) has garnered considerable
attention for its impressive performance, with ChatGPT emerging as a leading
AIGC model that produces high-quality responses across various applications,
including software development and maintenance. Despite its potential, the
misuse of ChatGPT poses significant concerns, especially in education and
safetycritical domains. Numerous AIGC detectors have been developed and
evaluated on natural language data. However, their performance on code-related
content generated by ChatGPT remains unexplored. To fill this gap, in this
paper, we present the first empirical study on evaluating existing AIGC
detectors in the software domain. We created a comprehensive dataset including
492.5K samples comprising code-related content produced by ChatGPT,
encompassing popular software activities like Q&A (115K), code summarization
(126K), and code generation (226.5K). We evaluated six AIGC detectors,
including three commercial and three open-source solutions, assessing their
performance on this dataset. Additionally, we conducted a human study to
understand human detection capabilities and compare them with the existing AIGC
detectors. Our results indicate that AIGC detectors demonstrate lower
performance on code-related data compared to natural language data. Fine-tuning
can enhance detector performance, especially for content within the same
domain; but generalization remains a challenge. The human evaluation reveals
that detection by humans is quite challenging.
| new_dataset | 0.959837 |
2304.05206 | Lu Han | Lu Han, Han-Jia Ye, De-Chuan Zhan | The Capacity and Robustness Trade-off: Revisiting the Channel
Independent Strategy for Multivariate Time Series Forecasting | under review | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multivariate time series data comprises various channels of variables. The
multivariate forecasting models need to capture the relationship between the
channels to accurately predict future values. However, recently, there has been
an emergence of methods that employ the Channel Independent (CI) strategy.
These methods view multivariate time series data as separate univariate time
series and disregard the correlation between channels. Surprisingly, our
empirical results have shown that models trained with the CI strategy
outperform those trained with the Channel Dependent (CD) strategy, usually by a
significant margin. Nevertheless, the reasons behind this phenomenon have not
yet been thoroughly explored in the literature. This paper provides
comprehensive empirical and theoretical analyses of the characteristics of
multivariate time series datasets and the CI/CD strategy. Our results conclude
that the CD approach has higher capacity but often lacks robustness to
accurately predict distributionally drifted time series. In contrast, the CI
approach trades capacity for robust prediction. Practical measures inspired by
these analyses are proposed to address the capacity and robustness dilemma,
including a modified CD method called Predict Residuals with Regularization
(PRReg) that can surpass the CI strategy. We hope our findings can raise
awareness among researchers about the characteristics of multivariate time
series and inspire the construction of better forecasting models.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 13:15:33 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Han",
"Lu",
""
],
[
"Ye",
"Han-Jia",
""
],
[
"Zhan",
"De-Chuan",
""
]
] | TITLE: The Capacity and Robustness Trade-off: Revisiting the Channel
Independent Strategy for Multivariate Time Series Forecasting
ABSTRACT: Multivariate time series data comprises various channels of variables. The
multivariate forecasting models need to capture the relationship between the
channels to accurately predict future values. However, recently, there has been
an emergence of methods that employ the Channel Independent (CI) strategy.
These methods view multivariate time series data as separate univariate time
series and disregard the correlation between channels. Surprisingly, our
empirical results have shown that models trained with the CI strategy
outperform those trained with the Channel Dependent (CD) strategy, usually by a
significant margin. Nevertheless, the reasons behind this phenomenon have not
yet been thoroughly explored in the literature. This paper provides
comprehensive empirical and theoretical analyses of the characteristics of
multivariate time series datasets and the CI/CD strategy. Our results conclude
that the CD approach has higher capacity but often lacks robustness to
accurately predict distributionally drifted time series. In contrast, the CI
approach trades capacity for robust prediction. Practical measures inspired by
these analyses are proposed to address the capacity and robustness dilemma,
including a modified CD method called Predict Residuals with Regularization
(PRReg) that can surpass the CI strategy. We hope our findings can raise
awareness among researchers about the characteristics of multivariate time
series and inspire the construction of better forecasting models.
| no_new_dataset | 0.948155 |
2304.05216 | Ensheng Shi | Ensheng Shi, Yanlin Wang, Hongyu Zhang, Lun Du, Shi Han, Dongmei
Zhang, Hongbin Sun | Towards Efficient Fine-tuning of Pre-trained Code Models: An
Experimental Study and Beyond | Accepted by ISSTA 2023 (The 32nd ACM SIGSOFT International Symposium
on Software Testing and Analysis) | null | null | null | cs.SE cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, fine-tuning pre-trained code models such as CodeBERT on downstream
tasks has achieved great success in many software testing and analysis tasks.
While effective and prevalent, fine-tuning the pre-trained parameters incurs a
large computational cost. In this paper, we conduct an extensive experimental
study to explore what happens to layer-wise pre-trained representations and
their encoded code knowledge during fine-tuning. We then propose efficient
alternatives to fine-tune the large pre-trained code model based on the above
findings. Our experimental study shows that (1) lexical, syntactic and
structural properties of source code are encoded in the lower, intermediate,
and higher layers, respectively, while the semantic property spans across the
entire model. (2) The process of fine-tuning preserves most of the code
properties. Specifically, the basic code properties captured by lower and
intermediate layers are still preserved during fine-tuning. Furthermore, we
find that only the representations of the top two layers change most during
fine-tuning for various downstream tasks. (3) Based on the above findings, we
propose Telly to efficiently fine-tune pre-trained code models via layer
freezing. The extensive experimental results on five various downstream tasks
demonstrate that training parameters and the corresponding time cost are
greatly reduced, while performances are similar or better. Replication package
including source code, datasets, and online Appendix is available at:
\url{https://github.com/DeepSoftwareAnalytics/Telly}.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 13:34:13 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Shi",
"Ensheng",
""
],
[
"Wang",
"Yanlin",
""
],
[
"Zhang",
"Hongyu",
""
],
[
"Du",
"Lun",
""
],
[
"Han",
"Shi",
""
],
[
"Zhang",
"Dongmei",
""
],
[
"Sun",
"Hongbin",
""
]
] | TITLE: Towards Efficient Fine-tuning of Pre-trained Code Models: An
Experimental Study and Beyond
ABSTRACT: Recently, fine-tuning pre-trained code models such as CodeBERT on downstream
tasks has achieved great success in many software testing and analysis tasks.
While effective and prevalent, fine-tuning the pre-trained parameters incurs a
large computational cost. In this paper, we conduct an extensive experimental
study to explore what happens to layer-wise pre-trained representations and
their encoded code knowledge during fine-tuning. We then propose efficient
alternatives to fine-tune the large pre-trained code model based on the above
findings. Our experimental study shows that (1) lexical, syntactic and
structural properties of source code are encoded in the lower, intermediate,
and higher layers, respectively, while the semantic property spans across the
entire model. (2) The process of fine-tuning preserves most of the code
properties. Specifically, the basic code properties captured by lower and
intermediate layers are still preserved during fine-tuning. Furthermore, we
find that only the representations of the top two layers change most during
fine-tuning for various downstream tasks. (3) Based on the above findings, we
propose Telly to efficiently fine-tune pre-trained code models via layer
freezing. The extensive experimental results on five various downstream tasks
demonstrate that training parameters and the corresponding time cost are
greatly reduced, while performances are similar or better. Replication package
including source code, datasets, and online Appendix is available at:
\url{https://github.com/DeepSoftwareAnalytics/Telly}.
| no_new_dataset | 0.946745 |
2304.05246 | William Jonas | W. Jonas, A. Abraham, L. Dreyfus-Schmidt | OpenAL: Evaluation and Interpretation of Active Learning Strategies | Published in NeurIPS 2022 Workshop on Human in the Loop Learning, 8
pages | null | null | null | cs.LG cs.AI cs.HC | http://creativecommons.org/licenses/by-sa/4.0/ | Despite the vast body of literature on Active Learning (AL), there is no
comprehensive and open benchmark allowing for efficient and simple comparison
of proposed samplers. Additionally, the variability in experimental settings
across the literature makes it difficult to choose a sampling strategy, which
is critical due to the one-off nature of AL experiments. To address those
limitations, we introduce OpenAL, a flexible and open-source framework to
easily run and compare sampling AL strategies on a collection of realistic
tasks. The proposed benchmark is augmented with interpretability metrics and
statistical analysis methods to understand when and why some samplers
outperform others. Last but not least, practitioners can easily extend the
benchmark by submitting their own AL samplers.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 14:35:14 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Jonas",
"W.",
""
],
[
"Abraham",
"A.",
""
],
[
"Dreyfus-Schmidt",
"L.",
""
]
] | TITLE: OpenAL: Evaluation and Interpretation of Active Learning Strategies
ABSTRACT: Despite the vast body of literature on Active Learning (AL), there is no
comprehensive and open benchmark allowing for efficient and simple comparison
of proposed samplers. Additionally, the variability in experimental settings
across the literature makes it difficult to choose a sampling strategy, which
is critical due to the one-off nature of AL experiments. To address those
limitations, we introduce OpenAL, a flexible and open-source framework to
easily run and compare sampling AL strategies on a collection of realistic
tasks. The proposed benchmark is augmented with interpretability metrics and
statistical analysis methods to understand when and why some samplers
outperform others. Last but not least, practitioners can easily extend the
benchmark by submitting their own AL samplers.
| no_new_dataset | 0.949716 |
2304.05255 | Chenshen Wu | Chenshen Wu and Joost van de Weijer | Density Map Distillation for Incremental Object Counting | Accepted at CVPR2023: Workshop on Continual Learning in Computer
Vision | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the problem of incremental learning for object counting, where
a method must learn to count a variety of object classes from a sequence of
datasets. A na\"ive approach to incremental object counting would suffer from
catastrophic forgetting, where it would suffer from a dramatic performance drop
on previous tasks. In this paper, we propose a new exemplar-free functional
regularization method, called Density Map Distillation (DMD). During training,
we introduce a new counter head for each task and introduce a distillation loss
to prevent forgetting of previous tasks. Additionally, we introduce a
cross-task adaptor that projects the features of the current backbone to the
previous backbone. This projector allows for the learning of new features while
the backbone retains the relevant features for previous tasks. Finally, we set
up experiments of incremental learning for counting new objects. Results
confirm that our method greatly reduces catastrophic forgetting and outperforms
existing methods.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 14:46:21 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wu",
"Chenshen",
""
],
[
"van de Weijer",
"Joost",
""
]
] | TITLE: Density Map Distillation for Incremental Object Counting
ABSTRACT: We investigate the problem of incremental learning for object counting, where
a method must learn to count a variety of object classes from a sequence of
datasets. A na\"ive approach to incremental object counting would suffer from
catastrophic forgetting, where it would suffer from a dramatic performance drop
on previous tasks. In this paper, we propose a new exemplar-free functional
regularization method, called Density Map Distillation (DMD). During training,
we introduce a new counter head for each task and introduce a distillation loss
to prevent forgetting of previous tasks. Additionally, we introduce a
cross-task adaptor that projects the features of the current backbone to the
previous backbone. This projector allows for the learning of new features while
the backbone retains the relevant features for previous tasks. Finally, we set
up experiments of incremental learning for counting new objects. Results
confirm that our method greatly reduces catastrophic forgetting and outperforms
existing methods.
| no_new_dataset | 0.947284 |
2304.05263 | Zizhuo Zhang | Zizhuo Zhang and Bang Wang | Prompt Learning for News Recommendation | 11 pages, 7 figures, 3 tables, SIGIR2023 conference | null | 10.1145/3539618.3591752 | null | cs.IR cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Some recent \textit{news recommendation} (NR) methods introduce a Pre-trained
Language Model (PLM) to encode news representation by following the vanilla
pre-train and fine-tune paradigm with carefully-designed
recommendation-specific neural networks and objective functions. Due to the
inconsistent task objective with that of PLM, we argue that their modeling
paradigm has not well exploited the abundant semantic information and
linguistic knowledge embedded in the pre-training process. Recently, the
pre-train, prompt, and predict paradigm, called \textit{prompt learning}, has
achieved many successes in natural language processing domain. In this paper,
we make the first trial of this new paradigm to develop a \textit{Prompt
Learning for News Recommendation} (Prompt4NR) framework, which transforms the
task of predicting whether a user would click a candidate news as a cloze-style
mask-prediction task. Specifically, we design a series of prompt templates,
including discrete, continuous, and hybrid templates, and construct their
corresponding answer spaces to examine the proposed Prompt4NR framework.
Furthermore, we use the prompt ensembling to integrate predictions from
multiple prompt templates. Extensive experiments on the MIND dataset validate
the effectiveness of our Prompt4NR with a set of new benchmark results.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 14:56:06 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Zhang",
"Zizhuo",
""
],
[
"Wang",
"Bang",
""
]
] | TITLE: Prompt Learning for News Recommendation
ABSTRACT: Some recent \textit{news recommendation} (NR) methods introduce a Pre-trained
Language Model (PLM) to encode news representation by following the vanilla
pre-train and fine-tune paradigm with carefully-designed
recommendation-specific neural networks and objective functions. Due to the
inconsistent task objective with that of PLM, we argue that their modeling
paradigm has not well exploited the abundant semantic information and
linguistic knowledge embedded in the pre-training process. Recently, the
pre-train, prompt, and predict paradigm, called \textit{prompt learning}, has
achieved many successes in natural language processing domain. In this paper,
we make the first trial of this new paradigm to develop a \textit{Prompt
Learning for News Recommendation} (Prompt4NR) framework, which transforms the
task of predicting whether a user would click a candidate news as a cloze-style
mask-prediction task. Specifically, we design a series of prompt templates,
including discrete, continuous, and hybrid templates, and construct their
corresponding answer spaces to examine the proposed Prompt4NR framework.
Furthermore, we use the prompt ensembling to integrate predictions from
multiple prompt templates. Extensive experiments on the MIND dataset validate
the effectiveness of our Prompt4NR with a set of new benchmark results.
| no_new_dataset | 0.942454 |
2304.05274 | Shao Yi Liaw | Shaoyi Liaw, Fan Huang, Fabricio Benevenuto, Haewoon Kwak, Jisun An | YouNICon: YouTube's CommuNIty of Conspiracy Videos | null | null | null | null | cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Conspiracy theories are widely propagated on social media. Among various
social media services, YouTube is one of the most influential sources of news
and entertainment. This paper seeks to develop a dataset, YOUNICON, to enable
researchers to perform conspiracy theory detection as well as classification of
videos with conspiracy theories into different topics. YOUNICON is a dataset
with a large collection of videos from suspicious channels that were identified
to contain conspiracy theories in a previous study (Ledwich and Zaitsev 2020).
Overall, YOUNICON will enable researchers to study trends in conspiracy
theories and understand how individuals can interact with the conspiracy theory
producing community or channel. Our data is available at:
https://doi.org/10.5281/zenodo.7466262.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 15:20:51 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Liaw",
"Shaoyi",
""
],
[
"Huang",
"Fan",
""
],
[
"Benevenuto",
"Fabricio",
""
],
[
"Kwak",
"Haewoon",
""
],
[
"An",
"Jisun",
""
]
] | TITLE: YouNICon: YouTube's CommuNIty of Conspiracy Videos
ABSTRACT: Conspiracy theories are widely propagated on social media. Among various
social media services, YouTube is one of the most influential sources of news
and entertainment. This paper seeks to develop a dataset, YOUNICON, to enable
researchers to perform conspiracy theory detection as well as classification of
videos with conspiracy theories into different topics. YOUNICON is a dataset
with a large collection of videos from suspicious channels that were identified
to contain conspiracy theories in a previous study (Ledwich and Zaitsev 2020).
Overall, YOUNICON will enable researchers to study trends in conspiracy
theories and understand how individuals can interact with the conspiracy theory
producing community or channel. Our data is available at:
https://doi.org/10.5281/zenodo.7466262.
| new_dataset | 0.964085 |
2304.05288 | Wenjin Wang | Wenjin Wang, Yunqing Hu, Qianglong Chen, Yin Zhang | Task Difficulty Aware Parameter Allocation & Regularization for Lifelong
Learning | Accepted by CVPR2023. Code is available at
https://github.com/WenjinW/PAR | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Parameter regularization or allocation methods are effective in overcoming
catastrophic forgetting in lifelong learning. However, they solve all tasks in
a sequence uniformly and ignore the differences in the learning difficulty of
different tasks. So parameter regularization methods face significant
forgetting when learning a new task very different from learned tasks, and
parameter allocation methods face unnecessary parameter overhead when learning
simple tasks. In this paper, we propose the Parameter Allocation &
Regularization (PAR), which adaptively select an appropriate strategy for each
task from parameter allocation and regularization based on its learning
difficulty. A task is easy for a model that has learned tasks related to it and
vice versa. We propose a divergence estimation method based on the
Nearest-Prototype distance to measure the task relatedness using only features
of the new task. Moreover, we propose a time-efficient relatedness-aware
sampling-based architecture search strategy to reduce the parameter overhead
for allocation. Experimental results on multiple benchmarks demonstrate that,
compared with SOTAs, our method is scalable and significantly reduces the
model's redundancy while improving the model's performance. Further qualitative
analysis indicates that PAR obtains reasonable task-relatedness.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 15:38:21 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wang",
"Wenjin",
""
],
[
"Hu",
"Yunqing",
""
],
[
"Chen",
"Qianglong",
""
],
[
"Zhang",
"Yin",
""
]
] | TITLE: Task Difficulty Aware Parameter Allocation & Regularization for Lifelong
Learning
ABSTRACT: Parameter regularization or allocation methods are effective in overcoming
catastrophic forgetting in lifelong learning. However, they solve all tasks in
a sequence uniformly and ignore the differences in the learning difficulty of
different tasks. So parameter regularization methods face significant
forgetting when learning a new task very different from learned tasks, and
parameter allocation methods face unnecessary parameter overhead when learning
simple tasks. In this paper, we propose the Parameter Allocation &
Regularization (PAR), which adaptively select an appropriate strategy for each
task from parameter allocation and regularization based on its learning
difficulty. A task is easy for a model that has learned tasks related to it and
vice versa. We propose a divergence estimation method based on the
Nearest-Prototype distance to measure the task relatedness using only features
of the new task. Moreover, we propose a time-efficient relatedness-aware
sampling-based architecture search strategy to reduce the parameter overhead
for allocation. Experimental results on multiple benchmarks demonstrate that,
compared with SOTAs, our method is scalable and significantly reduces the
model's redundancy while improving the model's performance. Further qualitative
analysis indicates that PAR obtains reasonable task-relatedness.
| no_new_dataset | 0.95018 |
2304.05293 | Ameya Thete | Daniel Murnane, Savannah Thais, Ameya Thete | Equivariant Graph Neural Networks for Charged Particle Tracking | Proceedings submission to ACAT 2022. 7 pages | null | null | null | physics.ins-det cs.LG hep-ex physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph neural networks (GNNs) have gained traction in high-energy physics
(HEP) for their potential to improve accuracy and scalability. However, their
resource-intensive nature and complex operations have motivated the development
of symmetry-equivariant architectures. In this work, we introduce EuclidNet, a
novel symmetry-equivariant GNN for charged particle tracking. EuclidNet
leverages the graph representation of collision events and enforces rotational
symmetry with respect to the detector's beamline axis, leading to a more
efficient model. We benchmark EuclidNet against the state-of-the-art
Interaction Network on the TrackML dataset, which simulates high-pileup
conditions expected at the High-Luminosity Large Hadron Collider (HL-LHC). Our
results show that EuclidNet achieves near-state-of-the-art performance at small
model scales (<1000 parameters), outperforming the non-equivariant benchmarks.
This study paves the way for future investigations into more resource-efficient
GNN models for particle tracking in HEP experiments.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 15:43:32 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Murnane",
"Daniel",
""
],
[
"Thais",
"Savannah",
""
],
[
"Thete",
"Ameya",
""
]
] | TITLE: Equivariant Graph Neural Networks for Charged Particle Tracking
ABSTRACT: Graph neural networks (GNNs) have gained traction in high-energy physics
(HEP) for their potential to improve accuracy and scalability. However, their
resource-intensive nature and complex operations have motivated the development
of symmetry-equivariant architectures. In this work, we introduce EuclidNet, a
novel symmetry-equivariant GNN for charged particle tracking. EuclidNet
leverages the graph representation of collision events and enforces rotational
symmetry with respect to the detector's beamline axis, leading to a more
efficient model. We benchmark EuclidNet against the state-of-the-art
Interaction Network on the TrackML dataset, which simulates high-pileup
conditions expected at the High-Luminosity Large Hadron Collider (HL-LHC). Our
results show that EuclidNet achieves near-state-of-the-art performance at small
model scales (<1000 parameters), outperforming the non-equivariant benchmarks.
This study paves the way for future investigations into more resource-efficient
GNN models for particle tracking in HEP experiments.
| no_new_dataset | 0.949248 |
2304.05295 | Hrishitva Patel | Hrishitva Patel | A Comprehensive Study on Object Detection Techniques in Unconstrained
Environments | 9 pages, 3 Figures, 2 Tables | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Object detection is a crucial task in computer vision that aims to identify
and localize objects in images or videos. The recent advancements in deep
learning and Convolutional Neural Networks (CNNs) have significantly improved
the performance of object detection techniques. This paper presents a
comprehensive study of object detection techniques in unconstrained
environments, including various challenges, datasets, and state-of-the-art
approaches. Additionally, we present a comparative analysis of the methods and
highlight their strengths and weaknesses. Finally, we provide some future
research directions to further improve object detection in unconstrained
environments.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 15:45:03 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Patel",
"Hrishitva",
""
]
] | TITLE: A Comprehensive Study on Object Detection Techniques in Unconstrained
Environments
ABSTRACT: Object detection is a crucial task in computer vision that aims to identify
and localize objects in images or videos. The recent advancements in deep
learning and Convolutional Neural Networks (CNNs) have significantly improved
the performance of object detection techniques. This paper presents a
comprehensive study of object detection techniques in unconstrained
environments, including various challenges, datasets, and state-of-the-art
approaches. Additionally, we present a comparative analysis of the methods and
highlight their strengths and weaknesses. Finally, we provide some future
research directions to further improve object detection in unconstrained
environments.
| no_new_dataset | 0.952131 |
2304.05296 | Ziyun Wang | Ziyun Wang, Kenneth Chaney, Kostas Daniilidis | EvAC3D: From Event-based Apparent Contours to 3D Models via Continuous
Visual Hulls | 16 pages, 8 figures, European Conference on Computer Vision (ECCV)
2022 | null | 10.1007/978-3-031-20071-7_17 | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | 3D reconstruction from multiple views is a successful computer vision field
with multiple deployments in applications. State of the art is based on
traditional RGB frames that enable optimization of photo-consistency cross
views. In this paper, we study the problem of 3D reconstruction from
event-cameras, motivated by the advantages of event-based cameras in terms of
low power and latency as well as by the biological evidence that eyes in nature
capture the same data and still perceive well 3D shape. The foundation of our
hypothesis that 3D reconstruction is feasible using events lies in the
information contained in the occluding contours and in the continuous scene
acquisition with events. We propose Apparent Contour Events (ACE), a novel
event-based representation that defines the geometry of the apparent contour of
an object. We represent ACE by a spatially and temporally continuous implicit
function defined in the event x-y-t space. Furthermore, we design a novel
continuous Voxel Carving algorithm enabled by the high temporal resolution of
the Apparent Contour Events. To evaluate the performance of the method, we
collect MOEC-3D, a 3D event dataset of a set of common real-world objects. We
demonstrate the ability of EvAC3D to reconstruct high-fidelity mesh surfaces
from real event sequences while allowing the refinement of the 3D
reconstruction for each individual event.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 15:46:16 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Wang",
"Ziyun",
""
],
[
"Chaney",
"Kenneth",
""
],
[
"Daniilidis",
"Kostas",
""
]
] | TITLE: EvAC3D: From Event-based Apparent Contours to 3D Models via Continuous
Visual Hulls
ABSTRACT: 3D reconstruction from multiple views is a successful computer vision field
with multiple deployments in applications. State of the art is based on
traditional RGB frames that enable optimization of photo-consistency cross
views. In this paper, we study the problem of 3D reconstruction from
event-cameras, motivated by the advantages of event-based cameras in terms of
low power and latency as well as by the biological evidence that eyes in nature
capture the same data and still perceive well 3D shape. The foundation of our
hypothesis that 3D reconstruction is feasible using events lies in the
information contained in the occluding contours and in the continuous scene
acquisition with events. We propose Apparent Contour Events (ACE), a novel
event-based representation that defines the geometry of the apparent contour of
an object. We represent ACE by a spatially and temporally continuous implicit
function defined in the event x-y-t space. Furthermore, we design a novel
continuous Voxel Carving algorithm enabled by the high temporal resolution of
the Apparent Contour Events. To evaluate the performance of the method, we
collect MOEC-3D, a 3D event dataset of a set of common real-world objects. We
demonstrate the ability of EvAC3D to reconstruct high-fidelity mesh surfaces
from real event sequences while allowing the refinement of the 3D
reconstruction for each individual event.
| new_dataset | 0.972571 |
2304.05310 | Qunxi Zhu | Qunxi Zhu, Yao Guo, Wei Lin | Neural Delay Differential Equations: System Reconstruction and Image
Classification | 11 pages, 12 figures. arXiv admin note: substantial text overlap with
arXiv:2102.10801 | null | null | null | cs.LG cs.AI math.DS nlin.CD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Neural Ordinary Differential Equations (NODEs), a framework of
continuous-depth neural networks, have been widely applied, showing exceptional
efficacy in coping with representative datasets. Recently, an augmented
framework has been developed to overcome some limitations that emerged in the
application of the original framework. In this paper, we propose a new class of
continuous-depth neural networks with delay, named Neural Delay Differential
Equations (NDDEs). To compute the corresponding gradients, we use the adjoint
sensitivity method to obtain the delayed dynamics of the adjoint. Differential
equations with delays are typically seen as dynamical systems of infinite
dimension that possess more fruitful dynamics. Compared to NODEs, NDDEs have a
stronger capacity of nonlinear representations. We use several illustrative
examples to demonstrate this outstanding capacity. Firstly, we successfully
model the delayed dynamics where the trajectories in the lower-dimensional
phase space could be mutually intersected and even chaotic in a model-free or
model-based manner. Traditional NODEs, without any argumentation, are not
directly applicable for such modeling. Secondly, we achieve lower loss and
higher accuracy not only for the data produced synthetically by complex models
but also for the CIFAR10, a well-known image dataset. Our results on the NDDEs
demonstrate that appropriately articulating the elements of dynamical systems
into the network design is truly beneficial in promoting network performance.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 16:09:28 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Zhu",
"Qunxi",
""
],
[
"Guo",
"Yao",
""
],
[
"Lin",
"Wei",
""
]
] | TITLE: Neural Delay Differential Equations: System Reconstruction and Image
Classification
ABSTRACT: Neural Ordinary Differential Equations (NODEs), a framework of
continuous-depth neural networks, have been widely applied, showing exceptional
efficacy in coping with representative datasets. Recently, an augmented
framework has been developed to overcome some limitations that emerged in the
application of the original framework. In this paper, we propose a new class of
continuous-depth neural networks with delay, named Neural Delay Differential
Equations (NDDEs). To compute the corresponding gradients, we use the adjoint
sensitivity method to obtain the delayed dynamics of the adjoint. Differential
equations with delays are typically seen as dynamical systems of infinite
dimension that possess more fruitful dynamics. Compared to NODEs, NDDEs have a
stronger capacity of nonlinear representations. We use several illustrative
examples to demonstrate this outstanding capacity. Firstly, we successfully
model the delayed dynamics where the trajectories in the lower-dimensional
phase space could be mutually intersected and even chaotic in a model-free or
model-based manner. Traditional NODEs, without any argumentation, are not
directly applicable for such modeling. Secondly, we achieve lower loss and
higher accuracy not only for the data produced synthetically by complex models
but also for the CIFAR10, a well-known image dataset. Our results on the NDDEs
demonstrate that appropriately articulating the elements of dynamical systems
into the network design is truly beneficial in promoting network performance.
| no_new_dataset | 0.941975 |
2304.05316 | Yunpeng Zhang | Yunpeng Zhang, Zheng Zhu, Dalong Du | OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy
Prediction | Code is available at https://github.com/zhangyp15/OccFormer | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The vision-based perception for autonomous driving has undergone a
transformation from the bird-eye-view (BEV) representations to the 3D semantic
occupancy. Compared with the BEV planes, the 3D semantic occupancy further
provides structural information along the vertical direction. This paper
presents OccFormer, a dual-path transformer network to effectively process the
3D volume for semantic occupancy prediction. OccFormer achieves a long-range,
dynamic, and efficient encoding of the camera-generated 3D voxel features. It
is obtained by decomposing the heavy 3D processing into the local and global
transformer pathways along the horizontal plane. For the occupancy decoder, we
adapt the vanilla Mask2Former for 3D semantic occupancy by proposing
preserve-pooling and class-guided sampling, which notably mitigate the sparsity
and class imbalance. Experimental results demonstrate that OccFormer
significantly outperforms existing methods for semantic scene completion on
SemanticKITTI dataset and for LiDAR semantic segmentation on nuScenes dataset.
Code is available at \url{https://github.com/zhangyp15/OccFormer}.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 16:15:50 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Zhang",
"Yunpeng",
""
],
[
"Zhu",
"Zheng",
""
],
[
"Du",
"Dalong",
""
]
] | TITLE: OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy
Prediction
ABSTRACT: The vision-based perception for autonomous driving has undergone a
transformation from the bird-eye-view (BEV) representations to the 3D semantic
occupancy. Compared with the BEV planes, the 3D semantic occupancy further
provides structural information along the vertical direction. This paper
presents OccFormer, a dual-path transformer network to effectively process the
3D volume for semantic occupancy prediction. OccFormer achieves a long-range,
dynamic, and efficient encoding of the camera-generated 3D voxel features. It
is obtained by decomposing the heavy 3D processing into the local and global
transformer pathways along the horizontal plane. For the occupancy decoder, we
adapt the vanilla Mask2Former for 3D semantic occupancy by proposing
preserve-pooling and class-guided sampling, which notably mitigate the sparsity
and class imbalance. Experimental results demonstrate that OccFormer
significantly outperforms existing methods for semantic scene completion on
SemanticKITTI dataset and for LiDAR semantic segmentation on nuScenes dataset.
Code is available at \url{https://github.com/zhangyp15/OccFormer}.
| no_new_dataset | 0.94868 |
2304.05326 | Stefan Riesenberger | Stefan Riesenberger and Christian Krieg | Towards Power Characterization of FPGA Architectures To Enable
Open-Source Power Estimation Using Micro-Benchmarks | Presented at the 3rd Workshop on Open-Source Design Automation
(OSDA), 2023 (arXiv:2303.18024) | null | null | OSDA/2023/05 | cs.AR | http://creativecommons.org/licenses/by/4.0/ | While in the past decade there has been significant progress in open-source
synthesis and verification tools and flows, one piece is still missing in the
open-source design automation ecosystem: a tool to estimate the power
consumption of a design on specific target technologies. We discuss a
work-in-progress method to characterize target technologies using generic
micro-benchmarks, whose results can be used to establish power models of these
target technologies. These models can further be used to predict the power
consumption of a design in a given use case scenario (which is currently out of
scope). We demonstrate our characterization method on the publicly documented
Lattice iCE40 FPGA technology, and discuss two approaches to generating
micro-benchmarks which consume power in the target device: simple lookup table
(LUT) instantiation, and a more sophisticated instantiation of ring
oscillators. We study three approaches to stimulate the implemented
micro-benchmarks in hardware: Verilog testbenches, micro-controller
testbenches, and pseudo-random linear-feedback-shift-register-(LFSR)-based
testing. We measure the power consumption of the stimulated target devices. Our
ultimate goal is to automate power measurements for technology
characterization; Currently, we manually measure the consumed power at three
shunt resistors using an oscilloscope. Preliminary results indicate that we are
able to induce variable power consumption in target devices; However, the
sensitivity of the power characterization is still too low to build expressive
power estimation models.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 16:33:20 GMT"
}
] | 2023-04-12T00:00:00 | [
[
"Riesenberger",
"Stefan",
""
],
[
"Krieg",
"Christian",
""
]
] | TITLE: Towards Power Characterization of FPGA Architectures To Enable
Open-Source Power Estimation Using Micro-Benchmarks
ABSTRACT: While in the past decade there has been significant progress in open-source
synthesis and verification tools and flows, one piece is still missing in the
open-source design automation ecosystem: a tool to estimate the power
consumption of a design on specific target technologies. We discuss a
work-in-progress method to characterize target technologies using generic
micro-benchmarks, whose results can be used to establish power models of these
target technologies. These models can further be used to predict the power
consumption of a design in a given use case scenario (which is currently out of
scope). We demonstrate our characterization method on the publicly documented
Lattice iCE40 FPGA technology, and discuss two approaches to generating
micro-benchmarks which consume power in the target device: simple lookup table
(LUT) instantiation, and a more sophisticated instantiation of ring
oscillators. We study three approaches to stimulate the implemented
micro-benchmarks in hardware: Verilog testbenches, micro-controller
testbenches, and pseudo-random linear-feedback-shift-register-(LFSR)-based
testing. We measure the power consumption of the stimulated target devices. Our
ultimate goal is to automate power measurements for technology
characterization; Currently, we manually measure the consumed power at three
shunt resistors using an oscilloscope. Preliminary results indicate that we are
able to induce variable power consumption in target devices; However, the
sensitivity of the power characterization is still too low to build expressive
power estimation models.
| no_new_dataset | 0.945298 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.