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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2303.08873 | Giorgis Georgakoudis | Giorgis Georgakoudis, Konstantinos Parasyris, Chunhua Liao, David
Beckingsale, Todd Gamblin, Bronis de Supinski | Machine Learning-Driven Adaptive OpenMP For Portable Performance on
Heterogeneous Systems | null | null | null | LLNL-CONF-833682 | cs.PL cs.DC cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Heterogeneity has become a mainstream architecture design choice for building
High Performance Computing systems. However, heterogeneity poses significant
challenges for achieving performance portability of execution. Adapting a
program to a new heterogeneous platform is laborious and requires developers to
manually explore a vast space of execution parameters. To address those
challenges, this paper proposes new extensions to OpenMP for autonomous,
machine learning-driven adaptation.
Our solution includes a set of novel language constructs, compiler
transformations, and runtime support. We propose a producer-consumer pattern to
flexibly define multiple, different variants of OpenMP code regions to enable
adaptation. Those regions are transparently profiled at runtime to autonomously
learn optimizing machine learning models that dynamically select the fastest
variant. Our approach significantly reduces users' efforts of programming
adaptive applications on heterogeneous architectures by leveraging machine
learning techniques and code generation capabilities of OpenMP compilation.
Using a complete reference implementation in Clang/LLVM we evaluate three
use-cases of adaptive CPU-GPU execution. Experiments with HPC proxy
applications and benchmarks demonstrate that the proposed adaptive OpenMP
extensions automatically choose the best performing code variants for various
adaptation possibilities, in several different heterogeneous platforms of CPUs
and GPUs.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 18:37:18 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Georgakoudis",
"Giorgis",
""
],
[
"Parasyris",
"Konstantinos",
""
],
[
"Liao",
"Chunhua",
""
],
[
"Beckingsale",
"David",
""
],
[
"Gamblin",
"Todd",
""
],
[
"de Supinski",
"Bronis",
""
]
] | TITLE: Machine Learning-Driven Adaptive OpenMP For Portable Performance on
Heterogeneous Systems
ABSTRACT: Heterogeneity has become a mainstream architecture design choice for building
High Performance Computing systems. However, heterogeneity poses significant
challenges for achieving performance portability of execution. Adapting a
program to a new heterogeneous platform is laborious and requires developers to
manually explore a vast space of execution parameters. To address those
challenges, this paper proposes new extensions to OpenMP for autonomous,
machine learning-driven adaptation.
Our solution includes a set of novel language constructs, compiler
transformations, and runtime support. We propose a producer-consumer pattern to
flexibly define multiple, different variants of OpenMP code regions to enable
adaptation. Those regions are transparently profiled at runtime to autonomously
learn optimizing machine learning models that dynamically select the fastest
variant. Our approach significantly reduces users' efforts of programming
adaptive applications on heterogeneous architectures by leveraging machine
learning techniques and code generation capabilities of OpenMP compilation.
Using a complete reference implementation in Clang/LLVM we evaluate three
use-cases of adaptive CPU-GPU execution. Experiments with HPC proxy
applications and benchmarks demonstrate that the proposed adaptive OpenMP
extensions automatically choose the best performing code variants for various
adaptation possibilities, in several different heterogeneous platforms of CPUs
and GPUs.
| no_new_dataset | 0.942348 |
2303.08883 | Riccardo Albertoni | Riccardo Albertoni, David Browning, Simon Cox, Alejandra N.
Gonzalez-Beltran, Andrea Perego, Peter Winstanley | The W3C Data Catalog Vocabulary, Version 2: Rationale, Design
Principles, and Uptake | null | null | null | null | cs.DL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | DCAT is an RDF vocabulary designed to facilitate interoperability between
data catalogs published on the Web. Since its first release in 2014 as a W3C
Recommendation, DCAT has seen a wide adoption across communities and domains,
particularly in conjunction with implementing the FAIR data principles (for
findable, accessible, interoperable and reusable data). These implementation
experiences, besides demonstrating the fitness of DCAT to meet its intended
purpose, helped identify existing issues and gaps. Moreover, over the last few
years, additional requirements emerged in data catalogs, given the increasing
practice of documenting not only datasets but also data services and APIs. This
paper illustrates the new version of DCAT, explaining the rationale behind its
main revisions and extensions, based on the collected use cases and
requirements, and outlines the issues yet to be addressed in future versions of
DCAT.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 18:59:53 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Albertoni",
"Riccardo",
""
],
[
"Browning",
"David",
""
],
[
"Cox",
"Simon",
""
],
[
"Gonzalez-Beltran",
"Alejandra N.",
""
],
[
"Perego",
"Andrea",
""
],
[
"Winstanley",
"Peter",
""
]
] | TITLE: The W3C Data Catalog Vocabulary, Version 2: Rationale, Design
Principles, and Uptake
ABSTRACT: DCAT is an RDF vocabulary designed to facilitate interoperability between
data catalogs published on the Web. Since its first release in 2014 as a W3C
Recommendation, DCAT has seen a wide adoption across communities and domains,
particularly in conjunction with implementing the FAIR data principles (for
findable, accessible, interoperable and reusable data). These implementation
experiences, besides demonstrating the fitness of DCAT to meet its intended
purpose, helped identify existing issues and gaps. Moreover, over the last few
years, additional requirements emerged in data catalogs, given the increasing
practice of documenting not only datasets but also data services and APIs. This
paper illustrates the new version of DCAT, explaining the rationale behind its
main revisions and extensions, based on the collected use cases and
requirements, and outlines the issues yet to be addressed in future versions of
DCAT.
| no_new_dataset | 0.949809 |
2303.08884 | Constantinos Siettos | Hector Vargas Alvarez, Gianluca Fabiani, Nikolaos Kazantzis,
Constantinos Siettos, Ioannis G. Kevrekidis | Discrete-Time Nonlinear Feedback Linearization via Physics-Informed
Machine Learning | null | null | null | null | cs.LG cs.NA math.DS math.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a physics-informed machine learning (PIML) scheme for the feedback
linearization of nonlinear discrete-time dynamical systems. The PIML finds the
nonlinear transformation law, thus ensuring stability via pole placement, in
one step. In order to facilitate convergence in the presence of steep gradients
in the nonlinear transformation law, we address a greedy-wise training
procedure. We assess the performance of the proposed PIML approach via a
benchmark nonlinear discrete map for which the feedback linearization
transformation law can be derived analytically; the example is characterized by
steep gradients, due to the presence of singularities, in the domain of
interest. We show that the proposed PIML outperforms, in terms of numerical
approximation accuracy, the traditional numerical implementation, which
involves the construction--and the solution in terms of the coefficients of a
power-series expansion--of a system of homological equations as well as the
implementation of the PIML in the entire domain, thus highlighting the
importance of continuation techniques in the training procedure of PIML.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 19:03:23 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Alvarez",
"Hector Vargas",
""
],
[
"Fabiani",
"Gianluca",
""
],
[
"Kazantzis",
"Nikolaos",
""
],
[
"Siettos",
"Constantinos",
""
],
[
"Kevrekidis",
"Ioannis G.",
""
]
] | TITLE: Discrete-Time Nonlinear Feedback Linearization via Physics-Informed
Machine Learning
ABSTRACT: We present a physics-informed machine learning (PIML) scheme for the feedback
linearization of nonlinear discrete-time dynamical systems. The PIML finds the
nonlinear transformation law, thus ensuring stability via pole placement, in
one step. In order to facilitate convergence in the presence of steep gradients
in the nonlinear transformation law, we address a greedy-wise training
procedure. We assess the performance of the proposed PIML approach via a
benchmark nonlinear discrete map for which the feedback linearization
transformation law can be derived analytically; the example is characterized by
steep gradients, due to the presence of singularities, in the domain of
interest. We show that the proposed PIML outperforms, in terms of numerical
approximation accuracy, the traditional numerical implementation, which
involves the construction--and the solution in terms of the coefficients of a
power-series expansion--of a system of homological equations as well as the
implementation of the PIML in the entire domain, thus highlighting the
importance of continuation techniques in the training procedure of PIML.
| no_new_dataset | 0.946843 |
2303.08889 | Bing He | Yingchen Ma, Bing He, Nathan Subrahmanian, Srijan Kumar | Characterizing and Predicting Social Correction on Twitter | Accepted for publication at: 15th ACM Web Science Conference 2023
(WebSci'23). Code and data at:
https://github.com/claws-lab/social-correction-twitter | null | null | null | cs.SI | http://creativecommons.org/licenses/by/4.0/ | Online misinformation has been a serious threat to public health and society.
Social media users are known to reply to misinformation posts with
counter-misinformation messages, which have been shown to be effective in
curbing the spread of misinformation. This is called social correction.
However, the characteristics of tweets that attract social correction versus
those that do not remain unknown. To close the gap, we focus on answering the
following two research questions: (1) ``Given a tweet, will it be countered by
other users?'', and (2) ``If yes, what will be the magnitude of countering
it?''. This exploration will help develop mechanisms to guide users'
misinformation correction efforts and to measure disparity across users who get
corrected. In this work, we first create a novel dataset with 690,047 pairs of
misinformation tweets and counter-misinformation replies. Then, stratified
analysis of tweet linguistic and engagement features as well as tweet posters'
user attributes are conducted to illustrate the factors that are significant in
determining whether a tweet will get countered. Finally, predictive classifiers
are created to predict the likelihood of a misinformation tweet to get
countered and the degree to which that tweet will be countered. The code and
data is accessible on https://github.com/claws-lab/social-correction-twitter.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 19:19:49 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Ma",
"Yingchen",
""
],
[
"He",
"Bing",
""
],
[
"Subrahmanian",
"Nathan",
""
],
[
"Kumar",
"Srijan",
""
]
] | TITLE: Characterizing and Predicting Social Correction on Twitter
ABSTRACT: Online misinformation has been a serious threat to public health and society.
Social media users are known to reply to misinformation posts with
counter-misinformation messages, which have been shown to be effective in
curbing the spread of misinformation. This is called social correction.
However, the characteristics of tweets that attract social correction versus
those that do not remain unknown. To close the gap, we focus on answering the
following two research questions: (1) ``Given a tweet, will it be countered by
other users?'', and (2) ``If yes, what will be the magnitude of countering
it?''. This exploration will help develop mechanisms to guide users'
misinformation correction efforts and to measure disparity across users who get
corrected. In this work, we first create a novel dataset with 690,047 pairs of
misinformation tweets and counter-misinformation replies. Then, stratified
analysis of tweet linguistic and engagement features as well as tweet posters'
user attributes are conducted to illustrate the factors that are significant in
determining whether a tweet will get countered. Finally, predictive classifiers
are created to predict the likelihood of a misinformation tweet to get
countered and the degree to which that tweet will be countered. The code and
data is accessible on https://github.com/claws-lab/social-correction-twitter.
| new_dataset | 0.960842 |
2303.08891 | Adar Kahana | Oded Ovadia, Adar Kahana, Panos Stinis, Eli Turkel, George Em
Karniadakis | ViTO: Vision Transformer-Operator | null | null | null | PNNL-SA-182861 | cs.CV cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | We combine vision transformers with operator learning to solve diverse
inverse problems described by partial differential equations (PDEs). Our
approach, named ViTO, combines a U-Net based architecture with a vision
transformer. We apply ViTO to solve inverse PDE problems of increasing
complexity, namely for the wave equation, the Navier-Stokes equations and the
Darcy equation. We focus on the more challenging case of super-resolution,
where the input dataset for the inverse problem is at a significantly coarser
resolution than the output. The results we obtain are comparable or exceed the
leading operator network benchmarks in terms of accuracy. Furthermore, ViTO`s
architecture has a small number of trainable parameters (less than 10% of the
leading competitor), resulting in a performance speed-up of over 5x when
averaged over the various test cases.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 19:24:14 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Ovadia",
"Oded",
""
],
[
"Kahana",
"Adar",
""
],
[
"Stinis",
"Panos",
""
],
[
"Turkel",
"Eli",
""
],
[
"Karniadakis",
"George Em",
""
]
] | TITLE: ViTO: Vision Transformer-Operator
ABSTRACT: We combine vision transformers with operator learning to solve diverse
inverse problems described by partial differential equations (PDEs). Our
approach, named ViTO, combines a U-Net based architecture with a vision
transformer. We apply ViTO to solve inverse PDE problems of increasing
complexity, namely for the wave equation, the Navier-Stokes equations and the
Darcy equation. We focus on the more challenging case of super-resolution,
where the input dataset for the inverse problem is at a significantly coarser
resolution than the output. The results we obtain are comparable or exceed the
leading operator network benchmarks in terms of accuracy. Furthermore, ViTO`s
architecture has a small number of trainable parameters (less than 10% of the
leading competitor), resulting in a performance speed-up of over 5x when
averaged over the various test cases.
| no_new_dataset | 0.952838 |
2303.08920 | Chenbin Pan | Chenbin Pan, Zhiqi Zhang, Senem Velipasalar, Yi Xu | EgoViT: Pyramid Video Transformer for Egocentric Action Recognition | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Capturing interaction of hands with objects is important to autonomously
detect human actions from egocentric videos. In this work, we present a pyramid
video transformer with a dynamic class token generator for egocentric action
recognition. Different from previous video transformers, which use the same
static embedding as the class token for diverse inputs, we propose a dynamic
class token generator that produces a class token for each input video by
analyzing the hand-object interaction and the related motion information. The
dynamic class token can diffuse such information to the entire model by
communicating with other informative tokens in the subsequent transformer
layers. With the dynamic class token, dissimilarity between videos can be more
prominent, which helps the model distinguish various inputs. In addition,
traditional video transformers explore temporal features globally, which
requires large amounts of computation. However, egocentric videos often have a
large amount of background scene transition, which causes discontinuities
across distant frames. In this case, blindly reducing the temporal sampling
rate will risk losing crucial information. Hence, we also propose a pyramid
architecture to hierarchically process the video from short-term high rate to
long-term low rate. With the proposed architecture, we significantly reduce the
computational cost as well as the memory requirement without sacrificing from
the model performance. We perform comparisons with different baseline video
transformers on the EPIC-KITCHENS-100 and EGTEA Gaze+ datasets. Both
quantitative and qualitative results show that the proposed model can
efficiently improve the performance for egocentric action recognition.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 20:33:50 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Pan",
"Chenbin",
""
],
[
"Zhang",
"Zhiqi",
""
],
[
"Velipasalar",
"Senem",
""
],
[
"Xu",
"Yi",
""
]
] | TITLE: EgoViT: Pyramid Video Transformer for Egocentric Action Recognition
ABSTRACT: Capturing interaction of hands with objects is important to autonomously
detect human actions from egocentric videos. In this work, we present a pyramid
video transformer with a dynamic class token generator for egocentric action
recognition. Different from previous video transformers, which use the same
static embedding as the class token for diverse inputs, we propose a dynamic
class token generator that produces a class token for each input video by
analyzing the hand-object interaction and the related motion information. The
dynamic class token can diffuse such information to the entire model by
communicating with other informative tokens in the subsequent transformer
layers. With the dynamic class token, dissimilarity between videos can be more
prominent, which helps the model distinguish various inputs. In addition,
traditional video transformers explore temporal features globally, which
requires large amounts of computation. However, egocentric videos often have a
large amount of background scene transition, which causes discontinuities
across distant frames. In this case, blindly reducing the temporal sampling
rate will risk losing crucial information. Hence, we also propose a pyramid
architecture to hierarchically process the video from short-term high rate to
long-term low rate. With the proposed architecture, we significantly reduce the
computational cost as well as the memory requirement without sacrificing from
the model performance. We perform comparisons with different baseline video
transformers on the EPIC-KITCHENS-100 and EGTEA Gaze+ datasets. Both
quantitative and qualitative results show that the proposed model can
efficiently improve the performance for egocentric action recognition.
| no_new_dataset | 0.935524 |
2303.08929 | Carlos Olarte | Jaime Arias, Kyungmin Bae, Carlos Olarte, Peter Csaba \"Olveczky,
Laure Petrucci, and Fredrik R{\o}mming | Symbolic Analysis and Parameter Synthesis for Time Petri Nets Using
Maude and SMT Solving | null | null | null | null | cs.LO | http://creativecommons.org/licenses/by/4.0/ | Parametric time Petri nets with inhibitor arcs (PITPNs) support flexibility
for timed systems by allowing parameters in firing bounds. In this paper we
present and prove correct a concrete and a symbolic rewriting logic semantics
for PITPNs. We show how this allows us to use Maude combined with SMT solving
to provide sound and complete formal analyses for PITPNs. We develop a new
general folding approach for symbolic reachability that terminates whenever the
parametric state-class graph of the PITPN is finite. We explain how almost all
formal analysis and parameter synthesis supported by the state-of-the-art PITPN
tool Rom\'eo can be done in Maude with SMT. In addition, we also support
analysis and parameter synthesis from parametric initial markings, as well as
full LTL model checking and analysis with user-defined execution strategies.
Experiments on three benchmarks show that our methods outperform Rom\'eo in
many cases.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 20:55:41 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Arias",
"Jaime",
""
],
[
"Bae",
"Kyungmin",
""
],
[
"Olarte",
"Carlos",
""
],
[
"Ölveczky",
"Peter Csaba",
""
],
[
"Petrucci",
"Laure",
""
],
[
"Rømming",
"Fredrik",
""
]
] | TITLE: Symbolic Analysis and Parameter Synthesis for Time Petri Nets Using
Maude and SMT Solving
ABSTRACT: Parametric time Petri nets with inhibitor arcs (PITPNs) support flexibility
for timed systems by allowing parameters in firing bounds. In this paper we
present and prove correct a concrete and a symbolic rewriting logic semantics
for PITPNs. We show how this allows us to use Maude combined with SMT solving
to provide sound and complete formal analyses for PITPNs. We develop a new
general folding approach for symbolic reachability that terminates whenever the
parametric state-class graph of the PITPN is finite. We explain how almost all
formal analysis and parameter synthesis supported by the state-of-the-art PITPN
tool Rom\'eo can be done in Maude with SMT. In addition, we also support
analysis and parameter synthesis from parametric initial markings, as well as
full LTL model checking and analysis with user-defined execution strategies.
Experiments on three benchmarks show that our methods outperform Rom\'eo in
many cases.
| no_new_dataset | 0.942242 |
2303.08934 | Wenxin Jiang | Wenxin Jiang, Nicholas Synovic, Purvish Jajal, Taylor R. Schorlemmer,
Arav Tewari, Bhavesh Pareek, George K. Thiruvathukal, James C. Davis | PTMTorrent: A Dataset for Mining Open-source Pre-trained Model Packages | 5 pages, 2 figures, Accepted to MSR'23 | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | Due to the cost of developing and training deep learning models from scratch,
machine learning engineers have begun to reuse pre-trained models (PTMs) and
fine-tune them for downstream tasks. PTM registries known as "model hubs"
support engineers in distributing and reusing deep learning models. PTM
packages include pre-trained weights, documentation, model architectures,
datasets, and metadata. Mining the information in PTM packages will enable the
discovery of engineering phenomena and tools to support software engineers.
However, accessing this information is difficult - there are many PTM
registries, and both the registries and the individual packages may have rate
limiting for accessing the data. We present an open-source dataset, PTMTorrent,
to facilitate the evaluation and understanding of PTM packages. This paper
describes the creation, structure, usage, and limitations of the dataset. The
dataset includes a snapshot of 5 model hubs and a total of 15,913 PTM packages.
These packages are represented in a uniform data schema for cross-hub mining.
We describe prior uses of this data and suggest research opportunities for
mining using our dataset. The PTMTorrent dataset (v1) is available at:
https://app.globus.org/file-manager?origin_id=55e17a6e-9d8f-11ed-a2a2-8383522b48d9&origin_path=%2F~%2F.
Our dataset generation tools are available on GitHub:
https://doi.org/10.5281/zenodo.7570357.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 21:01:31 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Jiang",
"Wenxin",
""
],
[
"Synovic",
"Nicholas",
""
],
[
"Jajal",
"Purvish",
""
],
[
"Schorlemmer",
"Taylor R.",
""
],
[
"Tewari",
"Arav",
""
],
[
"Pareek",
"Bhavesh",
""
],
[
"Thiruvathukal",
"George K.",
""
],
[
"Davis",
"James C.",
""
]
] | TITLE: PTMTorrent: A Dataset for Mining Open-source Pre-trained Model Packages
ABSTRACT: Due to the cost of developing and training deep learning models from scratch,
machine learning engineers have begun to reuse pre-trained models (PTMs) and
fine-tune them for downstream tasks. PTM registries known as "model hubs"
support engineers in distributing and reusing deep learning models. PTM
packages include pre-trained weights, documentation, model architectures,
datasets, and metadata. Mining the information in PTM packages will enable the
discovery of engineering phenomena and tools to support software engineers.
However, accessing this information is difficult - there are many PTM
registries, and both the registries and the individual packages may have rate
limiting for accessing the data. We present an open-source dataset, PTMTorrent,
to facilitate the evaluation and understanding of PTM packages. This paper
describes the creation, structure, usage, and limitations of the dataset. The
dataset includes a snapshot of 5 model hubs and a total of 15,913 PTM packages.
These packages are represented in a uniform data schema for cross-hub mining.
We describe prior uses of this data and suggest research opportunities for
mining using our dataset. The PTMTorrent dataset (v1) is available at:
https://app.globus.org/file-manager?origin_id=55e17a6e-9d8f-11ed-a2a2-8383522b48d9&origin_path=%2F~%2F.
Our dataset generation tools are available on GitHub:
https://doi.org/10.5281/zenodo.7570357.
| new_dataset | 0.974239 |
2303.08935 | Ahmad Bilal Asghar | Ahmad Bilal Asghar, Shreyas Sundaram, Stephen L. Smith | Multi-Robot Persistent Monitoring: Minimizing Latency and Number of
Robots with Recharging Constraints | 13 pages, 10 fiugres. arXiv admin note: substantial text overlap with
arXiv:1903.06105 | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we study multi-robot path planning for persistent monitoring
tasks. We consider the case where robots have a limited battery capacity with a
discharge time $D$. We represent the areas to be monitored as the vertices of a
weighted graph. For each vertex, there is a constraint on the maximum allowable
time between robot visits, called the latency. The objective is to find the
minimum number of robots that can satisfy these latency constraints while also
ensuring that the robots periodically charge at a recharging depot. The
decision version of this problem is known to be PSPACE-complete. We present a
$O(\frac{\log D}{\log \log D}\log \rho)$ approximation algorithm for the
problem where $\rho$ is the ratio of the maximum and the minimum latency
constraints. We also present an orienteering based heuristic to solve the
problem and show empirically that it typically provides higher quality
solutions than the approximation algorithm. We extend our results to provide an
algorithm for the problem of minimizing the maximum weighted latency given a
fixed number of robots. We evaluate our algorithms on large problem instances
in a patrolling scenario and in a wildfire monitoring application. We also
compare the algorithms with an existing solver on benchmark instances.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 21:03:24 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Asghar",
"Ahmad Bilal",
""
],
[
"Sundaram",
"Shreyas",
""
],
[
"Smith",
"Stephen L.",
""
]
] | TITLE: Multi-Robot Persistent Monitoring: Minimizing Latency and Number of
Robots with Recharging Constraints
ABSTRACT: In this paper we study multi-robot path planning for persistent monitoring
tasks. We consider the case where robots have a limited battery capacity with a
discharge time $D$. We represent the areas to be monitored as the vertices of a
weighted graph. For each vertex, there is a constraint on the maximum allowable
time between robot visits, called the latency. The objective is to find the
minimum number of robots that can satisfy these latency constraints while also
ensuring that the robots periodically charge at a recharging depot. The
decision version of this problem is known to be PSPACE-complete. We present a
$O(\frac{\log D}{\log \log D}\log \rho)$ approximation algorithm for the
problem where $\rho$ is the ratio of the maximum and the minimum latency
constraints. We also present an orienteering based heuristic to solve the
problem and show empirically that it typically provides higher quality
solutions than the approximation algorithm. We extend our results to provide an
algorithm for the problem of minimizing the maximum weighted latency given a
fixed number of robots. We evaluate our algorithms on large problem instances
in a patrolling scenario and in a wildfire monitoring application. We also
compare the algorithms with an existing solver on benchmark instances.
| no_new_dataset | 0.940953 |
2303.08970 | Haiguang Li | Haiguang Li, Trausti Thormundsson, Ivan Poupyrev, Nicholas Gillian | Gated Compression Layers for Efficient Always-On Models | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Mobile and embedded machine learning developers frequently have to compromise
between two inferior on-device deployment strategies: sacrifice accuracy and
aggressively shrink their models to run on dedicated low-power cores; or
sacrifice battery by running larger models on more powerful compute cores such
as neural processing units or the main application processor. In this paper, we
propose a novel Gated Compression layer that can be applied to transform
existing neural network architectures into Gated Neural Networks. Gated Neural
Networks have multiple properties that excel for on-device use cases that help
significantly reduce power, boost accuracy, and take advantage of heterogeneous
compute cores. We provide results across five public image and audio datasets
that demonstrate the proposed Gated Compression layer effectively stops up to
96% of negative samples, compresses 97% of positive samples, while maintaining
or improving model accuracy.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 22:46:22 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Li",
"Haiguang",
""
],
[
"Thormundsson",
"Trausti",
""
],
[
"Poupyrev",
"Ivan",
""
],
[
"Gillian",
"Nicholas",
""
]
] | TITLE: Gated Compression Layers for Efficient Always-On Models
ABSTRACT: Mobile and embedded machine learning developers frequently have to compromise
between two inferior on-device deployment strategies: sacrifice accuracy and
aggressively shrink their models to run on dedicated low-power cores; or
sacrifice battery by running larger models on more powerful compute cores such
as neural processing units or the main application processor. In this paper, we
propose a novel Gated Compression layer that can be applied to transform
existing neural network architectures into Gated Neural Networks. Gated Neural
Networks have multiple properties that excel for on-device use cases that help
significantly reduce power, boost accuracy, and take advantage of heterogeneous
compute cores. We provide results across five public image and audio datasets
that demonstrate the proposed Gated Compression layer effectively stops up to
96% of negative samples, compresses 97% of positive samples, while maintaining
or improving model accuracy.
| no_new_dataset | 0.94625 |
2303.08976 | Jacob Odg{\aa}rd T{\o}rring | Jacob O. T{\o}rring, Ben van Werkhoven, Filip Petrovic, Floris-Jan
Willemsen, Jir\'i Filipovic, Anne C. Elster | Towards a Benchmarking Suite for Kernel Tuners | null | null | null | null | cs.DC | http://creativecommons.org/licenses/by-sa/4.0/ | As computing system become more complex, it is becoming harder for
programmers to keep their codes optimized as the hardware gets updated.
Autotuners try to alleviate this by hiding as many architecture-based
optimization details as possible from the user, so that the code can be used
efficiently across different generations of systems.
In this article we introduce a new benchmark suite for evaluating the
performance of optimization algorithms used by modern autotuners targeting
GPUs. The suite contains tunable GPU kernels that are representative of
real-world applications, allowing for comparisons between optimization
algorithms and the examination of code optimization, search space difficulty,
and performance portability. Our framework facilitates easy integration of new
autotuners and benchmarks by defining a shared problem interface.
Our benchmark suite is evaluated based on five characteristics: convergence
rate, local minima centrality, optimal speedup, Permutation Feature Importance
(PFI), and performance portability. The results show that optimization
parameters greatly impact performance and the need for global optimization. The
importance of each parameter is consistent across GPU architectures, however,
the specific values need to be optimized for each architecture.
Our portability study highlights the crucial importance of autotuning each
application for a specific target architecture. The results reveal that simply
transferring the optimal configuration from one architecture to another can
result in a performance ranging from 58.5% to 99.9% of the optimal performance,
depending on the GPU architecture. This highlights the importance of autotuning
in modern computing systems and the value of our benchmark suite in
facilitating the study of optimization algorithms and their effectiveness in
achieving optimal performance for specific target architectures.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 22:55:40 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Tørring",
"Jacob O.",
""
],
[
"van Werkhoven",
"Ben",
""
],
[
"Petrovic",
"Filip",
""
],
[
"Willemsen",
"Floris-Jan",
""
],
[
"Filipovic",
"Jirí",
""
],
[
"Elster",
"Anne C.",
""
]
] | TITLE: Towards a Benchmarking Suite for Kernel Tuners
ABSTRACT: As computing system become more complex, it is becoming harder for
programmers to keep their codes optimized as the hardware gets updated.
Autotuners try to alleviate this by hiding as many architecture-based
optimization details as possible from the user, so that the code can be used
efficiently across different generations of systems.
In this article we introduce a new benchmark suite for evaluating the
performance of optimization algorithms used by modern autotuners targeting
GPUs. The suite contains tunable GPU kernels that are representative of
real-world applications, allowing for comparisons between optimization
algorithms and the examination of code optimization, search space difficulty,
and performance portability. Our framework facilitates easy integration of new
autotuners and benchmarks by defining a shared problem interface.
Our benchmark suite is evaluated based on five characteristics: convergence
rate, local minima centrality, optimal speedup, Permutation Feature Importance
(PFI), and performance portability. The results show that optimization
parameters greatly impact performance and the need for global optimization. The
importance of each parameter is consistent across GPU architectures, however,
the specific values need to be optimized for each architecture.
Our portability study highlights the crucial importance of autotuning each
application for a specific target architecture. The results reveal that simply
transferring the optimal configuration from one architecture to another can
result in a performance ranging from 58.5% to 99.9% of the optimal performance,
depending on the GPU architecture. This highlights the importance of autotuning
in modern computing systems and the value of our benchmark suite in
facilitating the study of optimization algorithms and their effectiveness in
achieving optimal performance for specific target architectures.
| no_new_dataset | 0.930046 |
2303.08985 | Lautaro Estienne | Lautaro Estienne, Matias Vera, Leonardo Rey Vega | Cross-domain Sentiment Classification in Spanish | null | null | 10.1109/ARGENCON55245.2022.9940056 | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sentiment Classification is a fundamental task in the field of Natural
Language Processing, and has very important academic and commercial
applications. It aims to automatically predict the degree of sentiment present
in a text that contains opinions and subjectivity at some level, like product
and movie reviews, or tweets. This can be really difficult to accomplish, in
part, because different domains of text contains different words and
expressions. In addition, this difficulty increases when text is written in a
non-English language due to the lack of databases and resources. As a
consequence, several cross-domain and cross-language techniques are often
applied to this task in order to improve the results. In this work we perform a
study on the ability of a classification system trained with a large database
of product reviews to generalize to different Spanish domains. Reviews were
collected from the MercadoLibre website from seven Latin American countries,
allowing the creation of a large and balanced dataset. Results suggest that
generalization across domains is feasible though very challenging when trained
with these product reviews, and can be improved by pre-training and fine-tuning
the classification model.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 23:11:30 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Estienne",
"Lautaro",
""
],
[
"Vera",
"Matias",
""
],
[
"Vega",
"Leonardo Rey",
""
]
] | TITLE: Cross-domain Sentiment Classification in Spanish
ABSTRACT: Sentiment Classification is a fundamental task in the field of Natural
Language Processing, and has very important academic and commercial
applications. It aims to automatically predict the degree of sentiment present
in a text that contains opinions and subjectivity at some level, like product
and movie reviews, or tweets. This can be really difficult to accomplish, in
part, because different domains of text contains different words and
expressions. In addition, this difficulty increases when text is written in a
non-English language due to the lack of databases and resources. As a
consequence, several cross-domain and cross-language techniques are often
applied to this task in order to improve the results. In this work we perform a
study on the ability of a classification system trained with a large database
of product reviews to generalize to different Spanish domains. Reviews were
collected from the MercadoLibre website from seven Latin American countries,
allowing the creation of a large and balanced dataset. Results suggest that
generalization across domains is feasible though very challenging when trained
with these product reviews, and can be improved by pre-training and fine-tuning
the classification model.
| no_new_dataset | 0.611382 |
2303.09005 | Wenjin Fu | Wenjin Fu, Yue Han, Jiangpeng He, Sriram Baireddy, Mridul Gupta,
Fengqing Zhu | Conditional Synthetic Food Image Generation | null | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generative Adversarial Networks (GAN) have been widely investigated for image
synthesis based on their powerful representation learning ability. In this
work, we explore the StyleGAN and its application of synthetic food image
generation. Despite the impressive performance of GAN for natural image
generation, food images suffer from high intra-class diversity and inter-class
similarity, resulting in overfitting and visual artifacts for synthetic images.
Therefore, we aim to explore the capability and improve the performance of GAN
methods for food image generation. Specifically, we first choose StyleGAN3 as
the baseline method to generate synthetic food images and analyze the
performance. Then, we identify two issues that can cause performance
degradation on food images during the training phase: (1) inter-class feature
entanglement during multi-food classes training and (2) loss of high-resolution
detail during image downsampling. To address both issues, we propose to train
one food category at a time to avoid feature entanglement and leverage image
patches cropped from high-resolution datasets to retain fine details. We
evaluate our method on the Food-101 dataset and show improved quality of
generated synthetic food images compared with the baseline. Finally, we
demonstrate the great potential of improving the performance of downstream
tasks, such as food image classification by including high-quality synthetic
training samples in the data augmentation.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 00:23:20 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Fu",
"Wenjin",
""
],
[
"Han",
"Yue",
""
],
[
"He",
"Jiangpeng",
""
],
[
"Baireddy",
"Sriram",
""
],
[
"Gupta",
"Mridul",
""
],
[
"Zhu",
"Fengqing",
""
]
] | TITLE: Conditional Synthetic Food Image Generation
ABSTRACT: Generative Adversarial Networks (GAN) have been widely investigated for image
synthesis based on their powerful representation learning ability. In this
work, we explore the StyleGAN and its application of synthetic food image
generation. Despite the impressive performance of GAN for natural image
generation, food images suffer from high intra-class diversity and inter-class
similarity, resulting in overfitting and visual artifacts for synthetic images.
Therefore, we aim to explore the capability and improve the performance of GAN
methods for food image generation. Specifically, we first choose StyleGAN3 as
the baseline method to generate synthetic food images and analyze the
performance. Then, we identify two issues that can cause performance
degradation on food images during the training phase: (1) inter-class feature
entanglement during multi-food classes training and (2) loss of high-resolution
detail during image downsampling. To address both issues, we propose to train
one food category at a time to avoid feature entanglement and leverage image
patches cropped from high-resolution datasets to retain fine details. We
evaluate our method on the Food-101 dataset and show improved quality of
generated synthetic food images compared with the baseline. Finally, we
demonstrate the great potential of improving the performance of downstream
tasks, such as food image classification by including high-quality synthetic
training samples in the data augmentation.
| no_new_dataset | 0.951774 |
2303.09007 | Yanhua Xu | Yanhua Xu | Machine Learning for Flow Cytometry Data Analysis | MSc thesis | null | null | null | cs.LG cs.AI q-bio.QM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Flow cytometry mainly used for detecting the characteristics of a number of
biochemical substances based on the expression of specific markers in cells. It
is particularly useful for detecting membrane surface receptors, antigens,
ions, or during DNA/RNA expression. Not only can it be employed as a biomedical
research tool for recognising distinctive types of cells in mixed populations,
but it can also be used as a diagnostic tool for classifying abnormal cell
populations connected with disease. Modern flow cytometers can rapidly analyse
tens of thousands of cells at the same time while also measuring multiple
parameters from a single cell. However, the rapid development of flow
cytometers makes it challenging for conventional analysis methods to interpret
flow cytometry data. Researchers need to be able to distinguish
interesting-looking cell populations manually in multi-dimensional data
collected from millions of cells. Thus, it is essential to find a robust
approach for analysing flow cytometry data automatically, specifically in
identifying cell populations automatically. This thesis mainly concerns
discover the potential shortcoming of current automated-gating algorithms in
both real datasets and synthetic datasets. Three representative automated
clustering algorithms are selected to be applied, compared and evaluated by
completely and partially automated gating. A subspace clustering ProClus also
implemented in this thesis. The performance of ProClus in flow cytometry is not
well, but it is still a useful algorithm to detect noise.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 00:43:46 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Xu",
"Yanhua",
""
]
] | TITLE: Machine Learning for Flow Cytometry Data Analysis
ABSTRACT: Flow cytometry mainly used for detecting the characteristics of a number of
biochemical substances based on the expression of specific markers in cells. It
is particularly useful for detecting membrane surface receptors, antigens,
ions, or during DNA/RNA expression. Not only can it be employed as a biomedical
research tool for recognising distinctive types of cells in mixed populations,
but it can also be used as a diagnostic tool for classifying abnormal cell
populations connected with disease. Modern flow cytometers can rapidly analyse
tens of thousands of cells at the same time while also measuring multiple
parameters from a single cell. However, the rapid development of flow
cytometers makes it challenging for conventional analysis methods to interpret
flow cytometry data. Researchers need to be able to distinguish
interesting-looking cell populations manually in multi-dimensional data
collected from millions of cells. Thus, it is essential to find a robust
approach for analysing flow cytometry data automatically, specifically in
identifying cell populations automatically. This thesis mainly concerns
discover the potential shortcoming of current automated-gating algorithms in
both real datasets and synthetic datasets. Three representative automated
clustering algorithms are selected to be applied, compared and evaluated by
completely and partially automated gating. A subspace clustering ProClus also
implemented in this thesis. The performance of ProClus in flow cytometry is not
well, but it is still a useful algorithm to detect noise.
| no_new_dataset | 0.947039 |
2303.09014 | Bhargavi Paranjape | Bhargavi Paranjape, Scott Lundberg, Sameer Singh, Hannaneh Hajishirzi,
Luke Zettlemoyer, Marco Tulio Ribeiro | ART: Automatic multi-step reasoning and tool-use for large language
models | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models (LLMs) can perform complex reasoning in few- and
zero-shot settings by generating intermediate chain of thought (CoT) reasoning
steps. Further, each reasoning step can rely on external tools to support
computation beyond the core LLM capabilities (e.g. search/running code). Prior
work on CoT prompting and tool use typically requires hand-crafting
task-specific demonstrations and carefully scripted interleaving of model
generations with tool use. We introduce Automatic Reasoning and Tool-use (ART),
a framework that uses frozen LLMs to automatically generate intermediate
reasoning steps as a program. Given a new task to solve, ART selects
demonstrations of multi-step reasoning and tool use from a task library. At
test time, ART seamlessly pauses generation whenever external tools are called,
and integrates their output before resuming generation. ART achieves a
substantial improvement over few-shot prompting and automatic CoT on unseen
tasks in the BigBench and MMLU benchmarks, and matches performance of
hand-crafted CoT prompts on a majority of these tasks. ART is also extensible,
and makes it easy for humans to improve performance by correcting errors in
task-specific programs or incorporating new tools, which we demonstrate by
drastically improving performance on select tasks with minimal human
intervention.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 01:04:45 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Paranjape",
"Bhargavi",
""
],
[
"Lundberg",
"Scott",
""
],
[
"Singh",
"Sameer",
""
],
[
"Hajishirzi",
"Hannaneh",
""
],
[
"Zettlemoyer",
"Luke",
""
],
[
"Ribeiro",
"Marco Tulio",
""
]
] | TITLE: ART: Automatic multi-step reasoning and tool-use for large language
models
ABSTRACT: Large language models (LLMs) can perform complex reasoning in few- and
zero-shot settings by generating intermediate chain of thought (CoT) reasoning
steps. Further, each reasoning step can rely on external tools to support
computation beyond the core LLM capabilities (e.g. search/running code). Prior
work on CoT prompting and tool use typically requires hand-crafting
task-specific demonstrations and carefully scripted interleaving of model
generations with tool use. We introduce Automatic Reasoning and Tool-use (ART),
a framework that uses frozen LLMs to automatically generate intermediate
reasoning steps as a program. Given a new task to solve, ART selects
demonstrations of multi-step reasoning and tool use from a task library. At
test time, ART seamlessly pauses generation whenever external tools are called,
and integrates their output before resuming generation. ART achieves a
substantial improvement over few-shot prompting and automatic CoT on unseen
tasks in the BigBench and MMLU benchmarks, and matches performance of
hand-crafted CoT prompts on a majority of these tasks. ART is also extensible,
and makes it easy for humans to improve performance by correcting errors in
task-specific programs or incorporating new tools, which we demonstrate by
drastically improving performance on select tasks with minimal human
intervention.
| no_new_dataset | 0.945851 |
2303.09024 | Arnab Bhattacharjee Mr | Arnab Bhattacharjee, Tapan K. Saha, Ashu Verma, Sukumar Mishra | DeeBBAA: A benchmark Deep Black Box Adversarial Attack against
Cyber-Physical Power Systems | null | null | null | null | cs.CR cs.SY eess.SY | http://creativecommons.org/licenses/by-sa/4.0/ | An increased energy demand, and environmental pressure to accommodate higher
levels of renewable energy and flexible loads like electric vehicles have led
to numerous smart transformations in the modern power systems. These
transformations make the cyber-physical power system highly susceptible to
cyber-adversaries targeting its numerous operations. In this work, a novel
black box adversarial attack strategy is proposed targeting the AC state
estimation operation of an unknown power system using historical data.
Specifically, false data is injected into the measurements obtained from a
small subset of the power system components which leads to significant
deviations in the state estimates. Experiments carried out on the IEEE 39 bus
and 118 bus test systems make it evident that the proposed strategy, called
DeeBBAA, can evade numerous conventional and state-of-the-art attack detection
mechanisms with very high probability.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 01:36:18 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Bhattacharjee",
"Arnab",
""
],
[
"Saha",
"Tapan K.",
""
],
[
"Verma",
"Ashu",
""
],
[
"Mishra",
"Sukumar",
""
]
] | TITLE: DeeBBAA: A benchmark Deep Black Box Adversarial Attack against
Cyber-Physical Power Systems
ABSTRACT: An increased energy demand, and environmental pressure to accommodate higher
levels of renewable energy and flexible loads like electric vehicles have led
to numerous smart transformations in the modern power systems. These
transformations make the cyber-physical power system highly susceptible to
cyber-adversaries targeting its numerous operations. In this work, a novel
black box adversarial attack strategy is proposed targeting the AC state
estimation operation of an unknown power system using historical data.
Specifically, false data is injected into the measurements obtained from a
small subset of the power system components which leads to significant
deviations in the state estimates. Experiments carried out on the IEEE 39 bus
and 118 bus test systems make it evident that the proposed strategy, called
DeeBBAA, can evade numerous conventional and state-of-the-art attack detection
mechanisms with very high probability.
| no_new_dataset | 0.946349 |
2303.09031 | Anthony Liu | Anthony Z. Liu, Lajanugen Logeswaran, Sungryull Sohn, Honglak Lee | A Picture is Worth a Thousand Words: Language Models Plan from Pixels | null | null | null | null | cs.CL cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Planning is an important capability of artificial agents that perform
long-horizon tasks in real-world environments. In this work, we explore the use
of pre-trained language models (PLMs) to reason about plan sequences from text
instructions in embodied visual environments. Prior PLM based approaches for
planning either assume observations are available in the form of text (e.g.,
provided by a captioning model), reason about plans from the instruction alone,
or incorporate information about the visual environment in limited ways (such
as a pre-trained affordance function). In contrast, we show that PLMs can
accurately plan even when observations are directly encoded as input prompts
for the PLM. We show that this simple approach outperforms prior approaches in
experiments on the ALFWorld and VirtualHome benchmarks.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 02:02:18 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Liu",
"Anthony Z.",
""
],
[
"Logeswaran",
"Lajanugen",
""
],
[
"Sohn",
"Sungryull",
""
],
[
"Lee",
"Honglak",
""
]
] | TITLE: A Picture is Worth a Thousand Words: Language Models Plan from Pixels
ABSTRACT: Planning is an important capability of artificial agents that perform
long-horizon tasks in real-world environments. In this work, we explore the use
of pre-trained language models (PLMs) to reason about plan sequences from text
instructions in embodied visual environments. Prior PLM based approaches for
planning either assume observations are available in the form of text (e.g.,
provided by a captioning model), reason about plans from the instruction alone,
or incorporate information about the visual environment in limited ways (such
as a pre-trained affordance function). In contrast, we show that PLMs can
accurately plan even when observations are directly encoded as input prompts
for the PLM. We show that this simple approach outperforms prior approaches in
experiments on the ALFWorld and VirtualHome benchmarks.
| no_new_dataset | 0.949856 |
2303.09046 | Andrew Peng | Andrew Peng, Jiangpeng He, Fengqing Zhu | Self-Supervised Visual Representation Learning on Food Images | Presented and published in EI 2023 Conference Proceedings | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Food image analysis is the groundwork for image-based dietary assessment,
which is the process of monitoring what kinds of food and how much energy is
consumed using captured food or eating scene images. Existing deep
learning-based methods learn the visual representation for downstream tasks
based on human annotation of each food image. However, most food images in real
life are obtained without labels, and data annotation requires plenty of time
and human effort, which is not feasible for real-world applications. To make
use of the vast amount of unlabeled images, many existing works focus on
unsupervised or self-supervised learning of visual representations directly
from unlabeled data. However, none of these existing works focus on food
images, which is more challenging than general objects due to its high
inter-class similarity and intra-class variance.
In this paper, we focus on the implementation and analysis of existing
representative self-supervised learning methods on food images. Specifically,
we first compare the performance of six selected self-supervised learning
models on the Food-101 dataset. Then we analyze the pros and cons of each
selected model when training on food data to identify the key factors that can
help improve the performance. Finally, we propose several ideas for future work
on self-supervised visual representation learning for food images.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 02:31:51 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Peng",
"Andrew",
""
],
[
"He",
"Jiangpeng",
""
],
[
"Zhu",
"Fengqing",
""
]
] | TITLE: Self-Supervised Visual Representation Learning on Food Images
ABSTRACT: Food image analysis is the groundwork for image-based dietary assessment,
which is the process of monitoring what kinds of food and how much energy is
consumed using captured food or eating scene images. Existing deep
learning-based methods learn the visual representation for downstream tasks
based on human annotation of each food image. However, most food images in real
life are obtained without labels, and data annotation requires plenty of time
and human effort, which is not feasible for real-world applications. To make
use of the vast amount of unlabeled images, many existing works focus on
unsupervised or self-supervised learning of visual representations directly
from unlabeled data. However, none of these existing works focus on food
images, which is more challenging than general objects due to its high
inter-class similarity and intra-class variance.
In this paper, we focus on the implementation and analysis of existing
representative self-supervised learning methods on food images. Specifically,
we first compare the performance of six selected self-supervised learning
models on the Food-101 dataset. Then we analyze the pros and cons of each
selected model when training on food data to identify the key factors that can
help improve the performance. Finally, we propose several ideas for future work
on self-supervised visual representation learning for food images.
| no_new_dataset | 0.949763 |
2303.09054 | Haruya Ishikawa | Haruya Ishikawa, Yoshimitsu Aoki | FindView: Precise Target View Localization Task for Look Around Agents | 19 pages, 7 figures, preprint, code available in
https://github.com/haruishi43/look_around | null | null | null | cs.CV cs.RO | http://creativecommons.org/licenses/by/4.0/ | With the increase in demands for service robots and automated inspection,
agents need to localize in its surrounding environment to achieve more natural
communication with humans by shared contexts. In this work, we propose a novel
but straightforward task of precise target view localization for look around
agents called the FindView task. This task imitates the movements of PTZ
cameras or user interfaces for 360 degree mediums, where the observer must
"look around" to find a view that exactly matches the target. To solve this
task, we introduce a rule-based agent that heuristically finds the optimal view
and a policy learning agent that employs reinforcement learning to learn by
interacting with the 360 degree scene. Through extensive evaluations and
benchmarks, we conclude that learned methods have many advantages, in
particular precise localization that is robust to corruption and can be easily
deployed in novel scenes.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 03:00:20 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Ishikawa",
"Haruya",
""
],
[
"Aoki",
"Yoshimitsu",
""
]
] | TITLE: FindView: Precise Target View Localization Task for Look Around Agents
ABSTRACT: With the increase in demands for service robots and automated inspection,
agents need to localize in its surrounding environment to achieve more natural
communication with humans by shared contexts. In this work, we propose a novel
but straightforward task of precise target view localization for look around
agents called the FindView task. This task imitates the movements of PTZ
cameras or user interfaces for 360 degree mediums, where the observer must
"look around" to find a view that exactly matches the target. To solve this
task, we introduce a rule-based agent that heuristically finds the optimal view
and a policy learning agent that employs reinforcement learning to learn by
interacting with the 360 degree scene. Through extensive evaluations and
benchmarks, we conclude that learned methods have many advantages, in
particular precise localization that is robust to corruption and can be easily
deployed in novel scenes.
| no_new_dataset | 0.947817 |
2303.09055 | Tuan Tang | Tuan N. Tang, Kwonyoung Kim, Kwanghoon Sohn | TemporalMaxer: Maximize Temporal Context with only Max Pooling for
Temporal Action Localization | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Temporal Action Localization (TAL) is a challenging task in video
understanding that aims to identify and localize actions within a video
sequence. Recent studies have emphasized the importance of applying long-term
temporal context modeling (TCM) blocks to the extracted video clip features
such as employing complex self-attention mechanisms. In this paper, we present
the simplest method ever to address this task and argue that the extracted
video clip features are already informative to achieve outstanding performance
without sophisticated architectures. To this end, we introduce TemporalMaxer,
which minimizes long-term temporal context modeling while maximizing
information from the extracted video clip features with a basic,
parameter-free, and local region operating max-pooling block. Picking out only
the most critical information for adjacent and local clip embeddings, this
block results in a more efficient TAL model. We demonstrate that TemporalMaxer
outperforms other state-of-the-art methods that utilize long-term TCM such as
self-attention on various TAL datasets while requiring significantly fewer
parameters and computational resources. The code for our approach is publicly
available at https://github.com/TuanTNG/TemporalMaxer
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 03:11:26 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Tang",
"Tuan N.",
""
],
[
"Kim",
"Kwonyoung",
""
],
[
"Sohn",
"Kwanghoon",
""
]
] | TITLE: TemporalMaxer: Maximize Temporal Context with only Max Pooling for
Temporal Action Localization
ABSTRACT: Temporal Action Localization (TAL) is a challenging task in video
understanding that aims to identify and localize actions within a video
sequence. Recent studies have emphasized the importance of applying long-term
temporal context modeling (TCM) blocks to the extracted video clip features
such as employing complex self-attention mechanisms. In this paper, we present
the simplest method ever to address this task and argue that the extracted
video clip features are already informative to achieve outstanding performance
without sophisticated architectures. To this end, we introduce TemporalMaxer,
which minimizes long-term temporal context modeling while maximizing
information from the extracted video clip features with a basic,
parameter-free, and local region operating max-pooling block. Picking out only
the most critical information for adjacent and local clip embeddings, this
block results in a more efficient TAL model. We demonstrate that TemporalMaxer
outperforms other state-of-the-art methods that utilize long-term TCM such as
self-attention on various TAL datasets while requiring significantly fewer
parameters and computational resources. The code for our approach is publicly
available at https://github.com/TuanTNG/TemporalMaxer
| no_new_dataset | 0.944125 |
2303.09057 | Hyun Joon Park | Hyun Joon Park, Seok Woo Yang, Jin Sob Kim, Wooseok Shin, Sung Won Han | TriAAN-VC: Triple Adaptive Attention Normalization for Any-to-Any Voice
Conversion | To appear in ICASSP 2023 | null | null | null | eess.AS cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Voice Conversion (VC) must be achieved while maintaining the content of the
source speech and representing the characteristics of the target speaker. The
existing methods do not simultaneously satisfy the above two aspects of VC, and
their conversion outputs suffer from a trade-off problem between maintaining
source contents and target characteristics. In this study, we propose Triple
Adaptive Attention Normalization VC (TriAAN-VC), comprising an encoder-decoder
and an attention-based adaptive normalization block, that can be applied to
non-parallel any-to-any VC. The proposed adaptive normalization block extracts
target speaker representations and achieves conversion while minimizing the
loss of the source content with siamese loss. We evaluated TriAAN-VC on the
VCTK dataset in terms of the maintenance of the source content and target
speaker similarity. Experimental results for one-shot VC suggest that TriAAN-VC
achieves state-of-the-art performance while mitigating the trade-off problem
encountered in the existing VC methods.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 03:13:58 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Park",
"Hyun Joon",
""
],
[
"Yang",
"Seok Woo",
""
],
[
"Kim",
"Jin Sob",
""
],
[
"Shin",
"Wooseok",
""
],
[
"Han",
"Sung Won",
""
]
] | TITLE: TriAAN-VC: Triple Adaptive Attention Normalization for Any-to-Any Voice
Conversion
ABSTRACT: Voice Conversion (VC) must be achieved while maintaining the content of the
source speech and representing the characteristics of the target speaker. The
existing methods do not simultaneously satisfy the above two aspects of VC, and
their conversion outputs suffer from a trade-off problem between maintaining
source contents and target characteristics. In this study, we propose Triple
Adaptive Attention Normalization VC (TriAAN-VC), comprising an encoder-decoder
and an attention-based adaptive normalization block, that can be applied to
non-parallel any-to-any VC. The proposed adaptive normalization block extracts
target speaker representations and achieves conversion while minimizing the
loss of the source content with siamese loss. We evaluated TriAAN-VC on the
VCTK dataset in terms of the maintenance of the source content and target
speaker similarity. Experimental results for one-shot VC suggest that TriAAN-VC
achieves state-of-the-art performance while mitigating the trade-off problem
encountered in the existing VC methods.
| no_new_dataset | 0.94743 |
2303.09061 | Liang Liu | Liang Liu, Boshen Zhang, Jiangning Zhang, Wuhao Zhang, Zhenye Gan,
Guanzhong Tian, Wenbing Zhu, Yabiao Wang, Chengjie Wang | MixTeacher: Mining Promising Labels with Mixed Scale Teacher for
Semi-Supervised Object Detection | Accepted by CVPR 2023. Implementation available:
https://github.com/lliuz/MixTeacher | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Scale variation across object instances remains a key challenge in object
detection task. Despite the remarkable progress made by modern detection
models, this challenge is particularly evident in the semi-supervised case.
While existing semi-supervised object detection methods rely on strict
conditions to filter high-quality pseudo labels from network predictions, we
observe that objects with extreme scale tend to have low confidence, resulting
in a lack of positive supervision for these objects. In this paper, we propose
a novel framework that addresses the scale variation problem by introducing a
mixed scale teacher to improve pseudo label generation and scale-invariant
learning. Additionally, we propose mining pseudo labels using score promotion
of predictions across scales, which benefits from better predictions from mixed
scale features. Our extensive experiments on MS COCO and PASCAL VOC benchmarks
under various semi-supervised settings demonstrate that our method achieves new
state-of-the-art performance. The code and models are available at
\url{https://github.com/lliuz/MixTeacher}.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 03:37:54 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Liu",
"Liang",
""
],
[
"Zhang",
"Boshen",
""
],
[
"Zhang",
"Jiangning",
""
],
[
"Zhang",
"Wuhao",
""
],
[
"Gan",
"Zhenye",
""
],
[
"Tian",
"Guanzhong",
""
],
[
"Zhu",
"Wenbing",
""
],
[
"Wang",
"Yabiao",
""
],
[
"Wang",
"Chengjie",
""
]
] | TITLE: MixTeacher: Mining Promising Labels with Mixed Scale Teacher for
Semi-Supervised Object Detection
ABSTRACT: Scale variation across object instances remains a key challenge in object
detection task. Despite the remarkable progress made by modern detection
models, this challenge is particularly evident in the semi-supervised case.
While existing semi-supervised object detection methods rely on strict
conditions to filter high-quality pseudo labels from network predictions, we
observe that objects with extreme scale tend to have low confidence, resulting
in a lack of positive supervision for these objects. In this paper, we propose
a novel framework that addresses the scale variation problem by introducing a
mixed scale teacher to improve pseudo label generation and scale-invariant
learning. Additionally, we propose mining pseudo labels using score promotion
of predictions across scales, which benefits from better predictions from mixed
scale features. Our extensive experiments on MS COCO and PASCAL VOC benchmarks
under various semi-supervised settings demonstrate that our method achieves new
state-of-the-art performance. The code and models are available at
\url{https://github.com/lliuz/MixTeacher}.
| no_new_dataset | 0.953794 |
2303.09088 | Yi Hong | Ankita Joshi and Yi Hong | MetaRegNet: Metamorphic Image Registration Using Flow-Driven Residual
Networks | 11 pages, 3 figures | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep learning based methods provide efficient solutions to medical image
registration, including the challenging problem of diffeomorphic image
registration. However, most methods register normal image pairs, facing
difficulty handling those with missing correspondences, e.g., in the presence
of pathology like tumors. We desire an efficient solution to jointly account
for spatial deformations and appearance changes in the pathological regions
where the correspondences are missing, i.e., finding a solution to metamorphic
image registration. Some approaches are proposed to tackle this problem, but
they cannot properly handle large pathological regions and deformations around
pathologies. In this paper, we propose a deep metamorphic image registration
network (MetaRegNet), which adopts time-varying flows to drive spatial
diffeomorphic deformations and generate intensity variations. We evaluate
MetaRegNet on two datasets, i.e., BraTS 2021 with brain tumors and 3D-IRCADb-01
with liver tumors, showing promising results in registering a healthy and tumor
image pair. The source code is available online.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 05:24:13 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Joshi",
"Ankita",
""
],
[
"Hong",
"Yi",
""
]
] | TITLE: MetaRegNet: Metamorphic Image Registration Using Flow-Driven Residual
Networks
ABSTRACT: Deep learning based methods provide efficient solutions to medical image
registration, including the challenging problem of diffeomorphic image
registration. However, most methods register normal image pairs, facing
difficulty handling those with missing correspondences, e.g., in the presence
of pathology like tumors. We desire an efficient solution to jointly account
for spatial deformations and appearance changes in the pathological regions
where the correspondences are missing, i.e., finding a solution to metamorphic
image registration. Some approaches are proposed to tackle this problem, but
they cannot properly handle large pathological regions and deformations around
pathologies. In this paper, we propose a deep metamorphic image registration
network (MetaRegNet), which adopts time-varying flows to drive spatial
diffeomorphic deformations and generate intensity variations. We evaluate
MetaRegNet on two datasets, i.e., BraTS 2021 with brain tumors and 3D-IRCADb-01
with liver tumors, showing promising results in registering a healthy and tumor
image pair. The source code is available online.
| no_new_dataset | 0.949295 |
2303.09114 | Shukang Yin | Shukang Yin, Shiwei Wu, Tong Xu, Shifeng Liu, Sirui Zhao, Enhong Chen | AU-aware graph convolutional network for Macro- and Micro-expression
spotting | Accepted by ICME-2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Automatic Micro-Expression (ME) spotting in long videos is a crucial step in
ME analysis but also a challenging task due to the short duration and low
intensity of MEs. When solving this problem, previous works generally lack in
considering the structures of human faces and the correspondence between
expressions and relevant facial muscles. To address this issue for better
performance of ME spotting, this paper seeks to extract finer spatial features
by modeling the relationships between facial Regions of Interest (ROIs).
Specifically, we propose a graph convolutional-based network, called
Action-Unit-aWare Graph Convolutional Network (AUW-GCN). Furthermore, to inject
prior information and to cope with the problem of small datasets, AU-related
statistics are encoded into the network. Comprehensive experiments show that
our results outperform baseline methods consistently and achieve new SOTA
performance in two benchmark datasets,CAS(ME)^2 and SAMM-LV. Our code is
available at https://github.com/xjtupanda/AUW-GCN.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 07:00:36 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Yin",
"Shukang",
""
],
[
"Wu",
"Shiwei",
""
],
[
"Xu",
"Tong",
""
],
[
"Liu",
"Shifeng",
""
],
[
"Zhao",
"Sirui",
""
],
[
"Chen",
"Enhong",
""
]
] | TITLE: AU-aware graph convolutional network for Macro- and Micro-expression
spotting
ABSTRACT: Automatic Micro-Expression (ME) spotting in long videos is a crucial step in
ME analysis but also a challenging task due to the short duration and low
intensity of MEs. When solving this problem, previous works generally lack in
considering the structures of human faces and the correspondence between
expressions and relevant facial muscles. To address this issue for better
performance of ME spotting, this paper seeks to extract finer spatial features
by modeling the relationships between facial Regions of Interest (ROIs).
Specifically, we propose a graph convolutional-based network, called
Action-Unit-aWare Graph Convolutional Network (AUW-GCN). Furthermore, to inject
prior information and to cope with the problem of small datasets, AU-related
statistics are encoded into the network. Comprehensive experiments show that
our results outperform baseline methods consistently and achieve new SOTA
performance in two benchmark datasets,CAS(ME)^2 and SAMM-LV. Our code is
available at https://github.com/xjtupanda/AUW-GCN.
| no_new_dataset | 0.952486 |
2303.09115 | Khiem Le | Cuong V. Nguyen, Khiem H. Le, Anh M. Tran, Quang H. Pham, Binh T.
Nguyen | Learning for Amalgamation: A Multi-Source Transfer Learning Framework
For Sentiment Classification | Information Sciences | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Transfer learning plays an essential role in Deep Learning, which can
remarkably improve the performance of the target domain, whose training data is
not sufficient. Our work explores beyond the common practice of transfer
learning with a single pre-trained model. We focus on the task of Vietnamese
sentiment classification and propose LIFA, a framework to learn a unified
embedding from several pre-trained models. We further propose two more LIFA
variants that encourage the pre-trained models to either cooperate or compete
with one another. Studying these variants sheds light on the success of LIFA by
showing that sharing knowledge among the models is more beneficial for transfer
learning. Moreover, we construct the AISIA-VN-Review-F dataset, the first
large-scale Vietnamese sentiment classification database. We conduct extensive
experiments on the AISIA-VN-Review-F and existing benchmarks to demonstrate the
efficacy of LIFA compared to other techniques. To contribute to the Vietnamese
NLP research, we publish our source code and datasets to the research community
upon acceptance.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 07:02:03 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Nguyen",
"Cuong V.",
""
],
[
"Le",
"Khiem H.",
""
],
[
"Tran",
"Anh M.",
""
],
[
"Pham",
"Quang H.",
""
],
[
"Nguyen",
"Binh T.",
""
]
] | TITLE: Learning for Amalgamation: A Multi-Source Transfer Learning Framework
For Sentiment Classification
ABSTRACT: Transfer learning plays an essential role in Deep Learning, which can
remarkably improve the performance of the target domain, whose training data is
not sufficient. Our work explores beyond the common practice of transfer
learning with a single pre-trained model. We focus on the task of Vietnamese
sentiment classification and propose LIFA, a framework to learn a unified
embedding from several pre-trained models. We further propose two more LIFA
variants that encourage the pre-trained models to either cooperate or compete
with one another. Studying these variants sheds light on the success of LIFA by
showing that sharing knowledge among the models is more beneficial for transfer
learning. Moreover, we construct the AISIA-VN-Review-F dataset, the first
large-scale Vietnamese sentiment classification database. We conduct extensive
experiments on the AISIA-VN-Review-F and existing benchmarks to demonstrate the
efficacy of LIFA compared to other techniques. To contribute to the Vietnamese
NLP research, we publish our source code and datasets to the research community
upon acceptance.
| new_dataset | 0.957118 |
2303.09135 | Nicolas E. Diaz Ferreyra PhD | Nicol\'as E. D\'iaz Ferreyra, Gautam Kishore Shahi, Catherine Tony,
Stefan Stieglitz and Riccardo Scandariato | Regret, Delete, (Do Not) Repeat: An Analysis of Self-Cleaning Practices
on Twitter After the Outbreak of the COVID-19 Pandemic | Accepted at CHI '23 Late Breaking Work (LBW) | null | 10.1145/3544549.3585583 | null | cs.SI cs.HC | http://creativecommons.org/licenses/by/4.0/ | During the outbreak of the COVID-19 pandemic, many people shared their
symptoms across Online Social Networks (OSNs) like Twitter, hoping for others'
advice or moral support. Prior studies have shown that those who disclose
health-related information across OSNs often tend to regret it and delete their
publications afterwards. Hence, deleted posts containing sensitive data can be
seen as manifestations of online regrets. In this work, we present an analysis
of deleted content on Twitter during the outbreak of the COVID-19 pandemic. For
this, we collected more than 3.67 million tweets describing COVID-19 symptoms
(e.g., fever, cough, and fatigue) posted between January and April 2020. We
observed that around 24% of the tweets containing personal pronouns were
deleted either by their authors or by the platform after one year. As a
practical application of the resulting dataset, we explored its suitability for
the automatic classification of regrettable content on Twitter.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 07:59:15 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Ferreyra",
"Nicolás E. Díaz",
""
],
[
"Shahi",
"Gautam Kishore",
""
],
[
"Tony",
"Catherine",
""
],
[
"Stieglitz",
"Stefan",
""
],
[
"Scandariato",
"Riccardo",
""
]
] | TITLE: Regret, Delete, (Do Not) Repeat: An Analysis of Self-Cleaning Practices
on Twitter After the Outbreak of the COVID-19 Pandemic
ABSTRACT: During the outbreak of the COVID-19 pandemic, many people shared their
symptoms across Online Social Networks (OSNs) like Twitter, hoping for others'
advice or moral support. Prior studies have shown that those who disclose
health-related information across OSNs often tend to regret it and delete their
publications afterwards. Hence, deleted posts containing sensitive data can be
seen as manifestations of online regrets. In this work, we present an analysis
of deleted content on Twitter during the outbreak of the COVID-19 pandemic. For
this, we collected more than 3.67 million tweets describing COVID-19 symptoms
(e.g., fever, cough, and fatigue) posted between January and April 2020. We
observed that around 24% of the tweets containing personal pronouns were
deleted either by their authors or by the platform after one year. As a
practical application of the resulting dataset, we explored its suitability for
the automatic classification of regrettable content on Twitter.
| no_new_dataset | 0.568865 |
2303.09152 | Xiaoyang Lyu | Xiaoyang Lyu, Peng Dai, Zizhang Li, Dongyu Yan, Yi Lin, Yifan Peng,
Xiaojuan Qi | Learning a Room with the Occ-SDF Hybrid: Signed Distance Function
Mingled with Occupancy Aids Scene Representation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Implicit neural rendering, which uses signed distance function (SDF)
representation with geometric priors (such as depth or surface normal), has led
to impressive progress in the surface reconstruction of large-scale scenes.
However, applying this method to reconstruct a room-level scene from images may
miss structures in low-intensity areas or small and thin objects. We conducted
experiments on three datasets to identify limitations of the original color
rendering loss and priors-embedded SDF scene representation.
We found that the color rendering loss results in optimization bias against
low-intensity areas, causing gradient vanishing and leaving these areas
unoptimized. To address this issue, we propose a feature-based color rendering
loss that utilizes non-zero feature values to bring back optimization signals.
Additionally, the SDF representation can be influenced by objects along a ray
path, disrupting the monotonic change of SDF values when a single object is
present. To counteract this, we explore using the occupancy representation,
which encodes each point separately and is unaffected by objects along a
querying ray. Our experimental results demonstrate that the joint forces of the
feature-based rendering loss and Occ-SDF hybrid representation scheme can
provide high-quality reconstruction results, especially in challenging
room-level scenarios. The code would be released.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 08:34:02 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Lyu",
"Xiaoyang",
""
],
[
"Dai",
"Peng",
""
],
[
"Li",
"Zizhang",
""
],
[
"Yan",
"Dongyu",
""
],
[
"Lin",
"Yi",
""
],
[
"Peng",
"Yifan",
""
],
[
"Qi",
"Xiaojuan",
""
]
] | TITLE: Learning a Room with the Occ-SDF Hybrid: Signed Distance Function
Mingled with Occupancy Aids Scene Representation
ABSTRACT: Implicit neural rendering, which uses signed distance function (SDF)
representation with geometric priors (such as depth or surface normal), has led
to impressive progress in the surface reconstruction of large-scale scenes.
However, applying this method to reconstruct a room-level scene from images may
miss structures in low-intensity areas or small and thin objects. We conducted
experiments on three datasets to identify limitations of the original color
rendering loss and priors-embedded SDF scene representation.
We found that the color rendering loss results in optimization bias against
low-intensity areas, causing gradient vanishing and leaving these areas
unoptimized. To address this issue, we propose a feature-based color rendering
loss that utilizes non-zero feature values to bring back optimization signals.
Additionally, the SDF representation can be influenced by objects along a ray
path, disrupting the monotonic change of SDF values when a single object is
present. To counteract this, we explore using the occupancy representation,
which encodes each point separately and is unaffected by objects along a
querying ray. Our experimental results demonstrate that the joint forces of the
feature-based rendering loss and Occ-SDF hybrid representation scheme can
provide high-quality reconstruction results, especially in challenging
room-level scenarios. The code would be released.
| no_new_dataset | 0.953319 |
2303.09153 | Zheyan Jin | Zheyan Jin, Shiqi Chen, Huajun Feng, Zhihai Xu, Qi Li, Yueting Chen | Reliable Image Dehazing by NeRF | 12pages, 8figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We present an image dehazing algorithm with high quality, wide application,
and no data training or prior needed. We analyze the defects of the original
dehazing model, and propose a new and reliable dehazing reconstruction and
dehazing model based on the combination of optical scattering model and
computer graphics lighting rendering model. Based on the new haze model and the
images obtained by the cameras, we can reconstruct the three-dimensional space,
accurately calculate the objects and haze in the space, and use the
transparency relationship of haze to perform accurate haze removal. To obtain a
3D simulation dataset we used the Unreal 5 computer graphics rendering engine.
In order to obtain real shot data in different scenes, we used fog generators,
array cameras, mobile phones, underwater cameras and drones to obtain haze
data. We use formula derivation, simulation data set and real shot data set
result experimental results to prove the feasibility of the new method.
Compared with various other methods, we are far ahead in terms of calculation
indicators (4 dB higher quality average scene), color remains more natural, and
the algorithm is more robust in different scenarios and best in the subjective
perception.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 08:34:03 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Jin",
"Zheyan",
""
],
[
"Chen",
"Shiqi",
""
],
[
"Feng",
"Huajun",
""
],
[
"Xu",
"Zhihai",
""
],
[
"Li",
"Qi",
""
],
[
"Chen",
"Yueting",
""
]
] | TITLE: Reliable Image Dehazing by NeRF
ABSTRACT: We present an image dehazing algorithm with high quality, wide application,
and no data training or prior needed. We analyze the defects of the original
dehazing model, and propose a new and reliable dehazing reconstruction and
dehazing model based on the combination of optical scattering model and
computer graphics lighting rendering model. Based on the new haze model and the
images obtained by the cameras, we can reconstruct the three-dimensional space,
accurately calculate the objects and haze in the space, and use the
transparency relationship of haze to perform accurate haze removal. To obtain a
3D simulation dataset we used the Unreal 5 computer graphics rendering engine.
In order to obtain real shot data in different scenes, we used fog generators,
array cameras, mobile phones, underwater cameras and drones to obtain haze
data. We use formula derivation, simulation data set and real shot data set
result experimental results to prove the feasibility of the new method.
Compared with various other methods, we are far ahead in terms of calculation
indicators (4 dB higher quality average scene), color remains more natural, and
the algorithm is more robust in different scenarios and best in the subjective
perception.
| no_new_dataset | 0.951549 |
2303.09166 | Imant Daunhawer | Imant Daunhawer, Alice Bizeul, Emanuele Palumbo, Alexander Marx, Julia
E. Vogt | Identifiability Results for Multimodal Contrastive Learning | ICLR 2023 camera-ready version | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by-sa/4.0/ | Contrastive learning is a cornerstone underlying recent progress in
multi-view and multimodal learning, e.g., in representation learning with
image/caption pairs. While its effectiveness is not yet fully understood, a
line of recent work reveals that contrastive learning can invert the data
generating process and recover ground truth latent factors shared between
views. In this work, we present new identifiability results for multimodal
contrastive learning, showing that it is possible to recover shared factors in
a more general setup than the multi-view setting studied previously.
Specifically, we distinguish between the multi-view setting with one generative
mechanism (e.g., multiple cameras of the same type) and the multimodal setting
that is characterized by distinct mechanisms (e.g., cameras and microphones).
Our work generalizes previous identifiability results by redefining the
generative process in terms of distinct mechanisms with modality-specific
latent variables. We prove that contrastive learning can block-identify latent
factors shared between modalities, even when there are nontrivial dependencies
between factors. We empirically verify our identifiability results with
numerical simulations and corroborate our findings on a complex multimodal
dataset of image/text pairs. Zooming out, our work provides a theoretical basis
for multimodal representation learning and explains in which settings
multimodal contrastive learning can be effective in practice.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 09:14:26 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Daunhawer",
"Imant",
""
],
[
"Bizeul",
"Alice",
""
],
[
"Palumbo",
"Emanuele",
""
],
[
"Marx",
"Alexander",
""
],
[
"Vogt",
"Julia E.",
""
]
] | TITLE: Identifiability Results for Multimodal Contrastive Learning
ABSTRACT: Contrastive learning is a cornerstone underlying recent progress in
multi-view and multimodal learning, e.g., in representation learning with
image/caption pairs. While its effectiveness is not yet fully understood, a
line of recent work reveals that contrastive learning can invert the data
generating process and recover ground truth latent factors shared between
views. In this work, we present new identifiability results for multimodal
contrastive learning, showing that it is possible to recover shared factors in
a more general setup than the multi-view setting studied previously.
Specifically, we distinguish between the multi-view setting with one generative
mechanism (e.g., multiple cameras of the same type) and the multimodal setting
that is characterized by distinct mechanisms (e.g., cameras and microphones).
Our work generalizes previous identifiability results by redefining the
generative process in terms of distinct mechanisms with modality-specific
latent variables. We prove that contrastive learning can block-identify latent
factors shared between modalities, even when there are nontrivial dependencies
between factors. We empirically verify our identifiability results with
numerical simulations and corroborate our findings on a complex multimodal
dataset of image/text pairs. Zooming out, our work provides a theoretical basis
for multimodal representation learning and explains in which settings
multimodal contrastive learning can be effective in practice.
| no_new_dataset | 0.950595 |
2303.09172 | Daniele Meli | Giulio Mazzi, Daniele Meli, Alberto Castellini, Alessandro Farinelli | Learning Logic Specifications for Soft Policy Guidance in POMCP | To appear in the Proceedings of 22nd International Conference on
Autonomous Agents and Multiagent Systems (AAMAS) 2023 | null | null | null | cs.AI cs.LG cs.LO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Partially Observable Monte Carlo Planning (POMCP) is an efficient solver for
Partially Observable Markov Decision Processes (POMDPs). It allows scaling to
large state spaces by computing an approximation of the optimal policy locally
and online, using a Monte Carlo Tree Search based strategy. However, POMCP
suffers from sparse reward function, namely, rewards achieved only when the
final goal is reached, particularly in environments with large state spaces and
long horizons. Recently, logic specifications have been integrated into POMCP
to guide exploration and to satisfy safety requirements. However, such
policy-related rules require manual definition by domain experts, especially in
real-world scenarios. In this paper, we use inductive logic programming to
learn logic specifications from traces of POMCP executions, i.e., sets of
belief-action pairs generated by the planner. Specifically, we learn rules
expressed in the paradigm of answer set programming. We then integrate them
inside POMCP to provide soft policy bias toward promising actions. In the
context of two benchmark scenarios, rocksample and battery, we show that the
integration of learned rules from small task instances can improve performance
with fewer Monte Carlo simulations and in larger task instances. We make our
modified version of POMCP publicly available at
https://github.com/GiuMaz/pomcp_clingo.git.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 09:37:10 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Mazzi",
"Giulio",
""
],
[
"Meli",
"Daniele",
""
],
[
"Castellini",
"Alberto",
""
],
[
"Farinelli",
"Alessandro",
""
]
] | TITLE: Learning Logic Specifications for Soft Policy Guidance in POMCP
ABSTRACT: Partially Observable Monte Carlo Planning (POMCP) is an efficient solver for
Partially Observable Markov Decision Processes (POMDPs). It allows scaling to
large state spaces by computing an approximation of the optimal policy locally
and online, using a Monte Carlo Tree Search based strategy. However, POMCP
suffers from sparse reward function, namely, rewards achieved only when the
final goal is reached, particularly in environments with large state spaces and
long horizons. Recently, logic specifications have been integrated into POMCP
to guide exploration and to satisfy safety requirements. However, such
policy-related rules require manual definition by domain experts, especially in
real-world scenarios. In this paper, we use inductive logic programming to
learn logic specifications from traces of POMCP executions, i.e., sets of
belief-action pairs generated by the planner. Specifically, we learn rules
expressed in the paradigm of answer set programming. We then integrate them
inside POMCP to provide soft policy bias toward promising actions. In the
context of two benchmark scenarios, rocksample and battery, we show that the
integration of learned rules from small task instances can improve performance
with fewer Monte Carlo simulations and in larger task instances. We make our
modified version of POMCP publicly available at
https://github.com/GiuMaz/pomcp_clingo.git.
| no_new_dataset | 0.946001 |
2303.09174 | Hao Liu | Hao Liu, Xin Li, Mingming Gong, Bing Liu, Yunfei Wu, Deqiang Jiang,
Yinsong Liu, Xing Sun | Grab What You Need: Rethinking Complex Table Structure Recognition with
Flexible Components Deliberation | Tech report | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, Table Structure Recognition (TSR) task, aiming at identifying table
structure into machine readable formats, has received increasing interest in
the community. While impressive success, most single table component-based
methods can not perform well on unregularized table cases distracted by not
only complicated inner structure but also exterior capture distortion. In this
paper, we raise it as Complex TSR problem, where the performance degeneration
of existing methods is attributable to their inefficient component usage and
redundant post-processing. To mitigate it, we shift our perspective from table
component extraction towards the efficient multiple components leverage, which
awaits further exploration in the field. Specifically, we propose a seminal
method, termed GrabTab, equipped with newly proposed Component Deliberator.
Thanks to its progressive deliberation mechanism, our GrabTab can flexibly
accommodate to most complex tables with reasonable components selected but
without complicated post-processing involved. Quantitative experimental results
on public benchmarks demonstrate that our method significantly outperforms the
state-of-the-arts, especially under more challenging scenes.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 09:37:38 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Liu",
"Hao",
""
],
[
"Li",
"Xin",
""
],
[
"Gong",
"Mingming",
""
],
[
"Liu",
"Bing",
""
],
[
"Wu",
"Yunfei",
""
],
[
"Jiang",
"Deqiang",
""
],
[
"Liu",
"Yinsong",
""
],
[
"Sun",
"Xing",
""
]
] | TITLE: Grab What You Need: Rethinking Complex Table Structure Recognition with
Flexible Components Deliberation
ABSTRACT: Recently, Table Structure Recognition (TSR) task, aiming at identifying table
structure into machine readable formats, has received increasing interest in
the community. While impressive success, most single table component-based
methods can not perform well on unregularized table cases distracted by not
only complicated inner structure but also exterior capture distortion. In this
paper, we raise it as Complex TSR problem, where the performance degeneration
of existing methods is attributable to their inefficient component usage and
redundant post-processing. To mitigate it, we shift our perspective from table
component extraction towards the efficient multiple components leverage, which
awaits further exploration in the field. Specifically, we propose a seminal
method, termed GrabTab, equipped with newly proposed Component Deliberator.
Thanks to its progressive deliberation mechanism, our GrabTab can flexibly
accommodate to most complex tables with reasonable components selected but
without complicated post-processing involved. Quantitative experimental results
on public benchmarks demonstrate that our method significantly outperforms the
state-of-the-arts, especially under more challenging scenes.
| no_new_dataset | 0.93784 |
2303.09187 | Zhongwei Qiu | Zhongwei Qiu, Yang Qiansheng, Jian Wang, Haocheng Feng, Junyu Han,
Errui Ding, Chang Xu, Dongmei Fu, Jingdong Wang | PSVT: End-to-End Multi-person 3D Pose and Shape Estimation with
Progressive Video Transformers | CVPR2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing methods of multi-person video 3D human Pose and Shape Estimation
(PSE) typically adopt a two-stage strategy, which first detects human instances
in each frame and then performs single-person PSE with temporal model. However,
the global spatio-temporal context among spatial instances can not be captured.
In this paper, we propose a new end-to-end multi-person 3D Pose and Shape
estimation framework with progressive Video Transformer, termed PSVT. In PSVT,
a spatio-temporal encoder (STE) captures the global feature dependencies among
spatial objects. Then, spatio-temporal pose decoder (STPD) and shape decoder
(STSD) capture the global dependencies between pose queries and feature tokens,
shape queries and feature tokens, respectively. To handle the variances of
objects as time proceeds, a novel scheme of progressive decoding is used to
update pose and shape queries at each frame. Besides, we propose a novel
pose-guided attention (PGA) for shape decoder to better predict shape
parameters. The two components strengthen the decoder of PSVT to improve
performance. Extensive experiments on the four datasets show that PSVT achieves
stage-of-the-art results.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 09:55:43 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Qiu",
"Zhongwei",
""
],
[
"Qiansheng",
"Yang",
""
],
[
"Wang",
"Jian",
""
],
[
"Feng",
"Haocheng",
""
],
[
"Han",
"Junyu",
""
],
[
"Ding",
"Errui",
""
],
[
"Xu",
"Chang",
""
],
[
"Fu",
"Dongmei",
""
],
[
"Wang",
"Jingdong",
""
]
] | TITLE: PSVT: End-to-End Multi-person 3D Pose and Shape Estimation with
Progressive Video Transformers
ABSTRACT: Existing methods of multi-person video 3D human Pose and Shape Estimation
(PSE) typically adopt a two-stage strategy, which first detects human instances
in each frame and then performs single-person PSE with temporal model. However,
the global spatio-temporal context among spatial instances can not be captured.
In this paper, we propose a new end-to-end multi-person 3D Pose and Shape
estimation framework with progressive Video Transformer, termed PSVT. In PSVT,
a spatio-temporal encoder (STE) captures the global feature dependencies among
spatial objects. Then, spatio-temporal pose decoder (STPD) and shape decoder
(STSD) capture the global dependencies between pose queries and feature tokens,
shape queries and feature tokens, respectively. To handle the variances of
objects as time proceeds, a novel scheme of progressive decoding is used to
update pose and shape queries at each frame. Besides, we propose a novel
pose-guided attention (PGA) for shape decoder to better predict shape
parameters. The two components strengthen the decoder of PSVT to improve
performance. Extensive experiments on the four datasets show that PSVT achieves
stage-of-the-art results.
| no_new_dataset | 0.944587 |
2303.09192 | Yuhang He | Yuhang He, Irving Fang, Yiming Li, Rushi Bhavesh Shah, Chen Feng | Metric-Free Exploration for Topological Mapping by Task and Motion
Imitation in Feature Space | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We propose DeepExplorer, a simple and lightweight metric-free exploration
method for topological mapping of unknown environments. It performs task and
motion planning (TAMP) entirely in image feature space. The task planner is a
recurrent network using the latest image observation sequence to hallucinate a
feature as the next-best exploration goal. The motion planner then utilizes the
current and the hallucinated features to generate an action taking the agent
towards that goal. The two planners are jointly trained via deeply-supervised
imitation learning from expert demonstrations. During exploration, we
iteratively call the two planners to predict the next action, and the
topological map is built by constantly appending the latest image observation
and action to the map and using visual place recognition (VPR) for loop
closing. The resulting topological map efficiently represents an environment's
connectivity and traversability, so it can be used for tasks such as visual
navigation. We show DeepExplorer's exploration efficiency and strong sim2sim
generalization capability on large-scale simulation datasets like Gibson and
MP3D. Its effectiveness is further validated via the image-goal navigation
performance on the resulting topological map. We further show its strong
zero-shot sim2real generalization capability in real-world experiments. The
source code is available at \url{https://ai4ce.github.io/DeepExplorer/}.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 10:04:41 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"He",
"Yuhang",
""
],
[
"Fang",
"Irving",
""
],
[
"Li",
"Yiming",
""
],
[
"Shah",
"Rushi Bhavesh",
""
],
[
"Feng",
"Chen",
""
]
] | TITLE: Metric-Free Exploration for Topological Mapping by Task and Motion
Imitation in Feature Space
ABSTRACT: We propose DeepExplorer, a simple and lightweight metric-free exploration
method for topological mapping of unknown environments. It performs task and
motion planning (TAMP) entirely in image feature space. The task planner is a
recurrent network using the latest image observation sequence to hallucinate a
feature as the next-best exploration goal. The motion planner then utilizes the
current and the hallucinated features to generate an action taking the agent
towards that goal. The two planners are jointly trained via deeply-supervised
imitation learning from expert demonstrations. During exploration, we
iteratively call the two planners to predict the next action, and the
topological map is built by constantly appending the latest image observation
and action to the map and using visual place recognition (VPR) for loop
closing. The resulting topological map efficiently represents an environment's
connectivity and traversability, so it can be used for tasks such as visual
navigation. We show DeepExplorer's exploration efficiency and strong sim2sim
generalization capability on large-scale simulation datasets like Gibson and
MP3D. Its effectiveness is further validated via the image-goal navigation
performance on the resulting topological map. We further show its strong
zero-shot sim2real generalization capability in real-world experiments. The
source code is available at \url{https://ai4ce.github.io/DeepExplorer/}.
| no_new_dataset | 0.95253 |
2303.09209 | Massimiliano Ronzani | Stefano Branchi, Andrei Buliga, Chiara Di Francescomarino, Chiara
Ghidini, Francesca Meneghello, Massimiliano Ronzani | Recommending the optimal policy by learning to act from temporal data | 10 pages, 5 figures | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Prescriptive Process Monitoring is a prominent problem in Process Mining,
which consists in identifying a set of actions to be recommended with the goal
of optimising a target measure of interest or Key Performance Indicator (KPI).
One challenge that makes this problem difficult is the need to provide
Prescriptive Process Monitoring techniques only based on temporally annotated
(process) execution data, stored in, so-called execution logs, due to the lack
of well crafted and human validated explicit models. In this paper we aim at
proposing an AI based approach that learns, by means of Reinforcement Learning
(RL), an optimal policy (almost) only from the observation of past executions
and recommends the best activities to carry on for optimizing a KPI of
interest. This is achieved first by learning a Markov Decision Process for the
specific KPIs from data, and then by using RL training to learn the optimal
policy. The approach is validated on real and synthetic datasets and compared
with off-policy Deep RL approaches. The ability of our approach to compare
with, and often overcome, Deep RL approaches provides a contribution towards
the exploitation of white box RL techniques in scenarios where only temporal
execution data are available.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 10:30:36 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Branchi",
"Stefano",
""
],
[
"Buliga",
"Andrei",
""
],
[
"Di Francescomarino",
"Chiara",
""
],
[
"Ghidini",
"Chiara",
""
],
[
"Meneghello",
"Francesca",
""
],
[
"Ronzani",
"Massimiliano",
""
]
] | TITLE: Recommending the optimal policy by learning to act from temporal data
ABSTRACT: Prescriptive Process Monitoring is a prominent problem in Process Mining,
which consists in identifying a set of actions to be recommended with the goal
of optimising a target measure of interest or Key Performance Indicator (KPI).
One challenge that makes this problem difficult is the need to provide
Prescriptive Process Monitoring techniques only based on temporally annotated
(process) execution data, stored in, so-called execution logs, due to the lack
of well crafted and human validated explicit models. In this paper we aim at
proposing an AI based approach that learns, by means of Reinforcement Learning
(RL), an optimal policy (almost) only from the observation of past executions
and recommends the best activities to carry on for optimizing a KPI of
interest. This is achieved first by learning a Markov Decision Process for the
specific KPIs from data, and then by using RL training to learn the optimal
policy. The approach is validated on real and synthetic datasets and compared
with off-policy Deep RL approaches. The ability of our approach to compare
with, and often overcome, Deep RL approaches provides a contribution towards
the exploitation of white box RL techniques in scenarios where only temporal
execution data are available.
| no_new_dataset | 0.943971 |
2303.09212 | Mingyue Zhao | Mingyue Zhao, Shang Zhao, Quan Quan, Li Fan, Xiaolan Qiu, Shiyuan Liu,
and S.Kevin Zhou | GDDS: Pulmonary Bronchioles Segmentation with Group Deep Dense
Supervision | Submitted to MICCAI2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Airway segmentation, especially bronchioles segmentation, is an important but
challenging task because distal bronchus are sparsely distributed and of a fine
scale. Existing neural networks usually exploit sparse topology to learn the
connectivity of bronchioles and inefficient shallow features to capture such
high-frequency information, leading to the breakage or missed detection of
individual thin branches. To address these problems, we contribute a new
bronchial segmentation method based on Group Deep Dense Supervision (GDDS) that
emphasizes fine-scale bronchioles segmentation in a simple-but-effective
manner. First, Deep Dense Supervision (DDS) is proposed by constructing local
dense topology skillfully and implementing dense topological learning on a
specific shallow feature layer. GDDS further empowers the shallow features with
better perception ability to detect bronchioles, even the ones that are not
easily discernible to the naked eye. Extensive experiments on the BAS benchmark
dataset have shown that our method promotes the network to have a high
sensitivity in capturing fine-scale branches and outperforms state-of-the-art
methods by a large margin (+12.8 % in BD and +8.8 % in TD) while only
introducing a small number of extra parameters.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 10:35:32 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Zhao",
"Mingyue",
""
],
[
"Zhao",
"Shang",
""
],
[
"Quan",
"Quan",
""
],
[
"Fan",
"Li",
""
],
[
"Qiu",
"Xiaolan",
""
],
[
"Liu",
"Shiyuan",
""
],
[
"Zhou",
"S. Kevin",
""
]
] | TITLE: GDDS: Pulmonary Bronchioles Segmentation with Group Deep Dense
Supervision
ABSTRACT: Airway segmentation, especially bronchioles segmentation, is an important but
challenging task because distal bronchus are sparsely distributed and of a fine
scale. Existing neural networks usually exploit sparse topology to learn the
connectivity of bronchioles and inefficient shallow features to capture such
high-frequency information, leading to the breakage or missed detection of
individual thin branches. To address these problems, we contribute a new
bronchial segmentation method based on Group Deep Dense Supervision (GDDS) that
emphasizes fine-scale bronchioles segmentation in a simple-but-effective
manner. First, Deep Dense Supervision (DDS) is proposed by constructing local
dense topology skillfully and implementing dense topological learning on a
specific shallow feature layer. GDDS further empowers the shallow features with
better perception ability to detect bronchioles, even the ones that are not
easily discernible to the naked eye. Extensive experiments on the BAS benchmark
dataset have shown that our method promotes the network to have a high
sensitivity in capturing fine-scale branches and outperforms state-of-the-art
methods by a large margin (+12.8 % in BD and +8.8 % in TD) while only
introducing a small number of extra parameters.
| no_new_dataset | 0.947914 |
2303.09252 | Jiayi Lin | Jiayi Lin, Shaogang Gong | GridCLIP: One-Stage Object Detection by Grid-Level CLIP Representation
Learning | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A vision-language foundation model pretrained on very large-scale image-text
paired data has the potential to provide generalizable knowledge representation
for downstream visual recognition and detection tasks, especially on
supplementing the undersampled categories in downstream model training. Recent
studies utilizing CLIP for object detection have shown that a two-stage
detector design typically outperforms a one-stage detector, while requiring
more expensive training resources and longer inference time. In this work, we
propose a one-stage detector GridCLIP that narrows its performance gap to those
of two-stage detectors, with approximately 43 and 5 times faster than its
two-stage counterpart (ViLD) in the training and test process respectively.
GridCLIP learns grid-level representations to adapt to the intrinsic principle
of one-stage detection learning by expanding the conventional CLIP image-text
holistic mapping to a more fine-grained, grid-text alignment. This differs from
the region-text mapping in two-stage detectors that apply CLIP directly by
treating regions as images. Specifically, GridCLIP performs Grid-level
Alignment to adapt the CLIP image-level representations to grid-level
representations by aligning to CLIP category representations to learn the
annotated (especially frequent) categories. To learn generalizable visual
representations of broader categories, especially undersampled ones, we perform
Image-level Alignment during training to propagate broad pre-learned categories
in the CLIP image encoder from the image-level to the grid-level
representations. Experiments show that the learned CLIP-based grid-level
representations boost the performance of undersampled (infrequent and novel)
categories, reaching comparable detection performance on the LVIS benchmark.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 12:06:02 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Lin",
"Jiayi",
""
],
[
"Gong",
"Shaogang",
""
]
] | TITLE: GridCLIP: One-Stage Object Detection by Grid-Level CLIP Representation
Learning
ABSTRACT: A vision-language foundation model pretrained on very large-scale image-text
paired data has the potential to provide generalizable knowledge representation
for downstream visual recognition and detection tasks, especially on
supplementing the undersampled categories in downstream model training. Recent
studies utilizing CLIP for object detection have shown that a two-stage
detector design typically outperforms a one-stage detector, while requiring
more expensive training resources and longer inference time. In this work, we
propose a one-stage detector GridCLIP that narrows its performance gap to those
of two-stage detectors, with approximately 43 and 5 times faster than its
two-stage counterpart (ViLD) in the training and test process respectively.
GridCLIP learns grid-level representations to adapt to the intrinsic principle
of one-stage detection learning by expanding the conventional CLIP image-text
holistic mapping to a more fine-grained, grid-text alignment. This differs from
the region-text mapping in two-stage detectors that apply CLIP directly by
treating regions as images. Specifically, GridCLIP performs Grid-level
Alignment to adapt the CLIP image-level representations to grid-level
representations by aligning to CLIP category representations to learn the
annotated (especially frequent) categories. To learn generalizable visual
representations of broader categories, especially undersampled ones, we perform
Image-level Alignment during training to propagate broad pre-learned categories
in the CLIP image encoder from the image-level to the grid-level
representations. Experiments show that the learned CLIP-based grid-level
representations boost the performance of undersampled (infrequent and novel)
categories, reaching comparable detection performance on the LVIS benchmark.
| no_new_dataset | 0.950411 |
2303.09253 | Wenjian Wang | Wenjian Wang, Lijuan Duan, Yuxi Wang, Junsong Fan, Zhi Gong, Zhaoxiang
Zhang | A Survey of Deep Visual Cross-Domain Few-Shot Learning | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Few-Shot transfer learning has become a major focus of research as it allows
recognition of new classes with limited labeled data. While it is assumed that
train and test data have the same data distribution, this is often not the case
in real-world applications. This leads to decreased model transfer effects when
the new class distribution differs significantly from the learned classes.
Research into Cross-Domain Few-Shot (CDFS) has emerged to address this issue,
forming a more challenging and realistic setting. In this survey, we provide a
detailed taxonomy of CDFS from the problem setting and corresponding solutions
view. We summarise the existing CDFS network architectures and discuss the
solution ideas for each direction the taxonomy indicates. Furthermore, we
introduce various CDFS downstream applications and outline classification,
detection, and segmentation benchmarks and corresponding standards for
evaluation. We also discuss the challenges of CDFS research and explore
potential directions for future investigation. Through this review, we aim to
provide comprehensive guidance on CDFS research, enabling researchers to gain
insight into the state-of-the-art while allowing them to build upon existing
solutions to develop their own CDFS models.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 12:06:59 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Wang",
"Wenjian",
""
],
[
"Duan",
"Lijuan",
""
],
[
"Wang",
"Yuxi",
""
],
[
"Fan",
"Junsong",
""
],
[
"Gong",
"Zhi",
""
],
[
"Zhang",
"Zhaoxiang",
""
]
] | TITLE: A Survey of Deep Visual Cross-Domain Few-Shot Learning
ABSTRACT: Few-Shot transfer learning has become a major focus of research as it allows
recognition of new classes with limited labeled data. While it is assumed that
train and test data have the same data distribution, this is often not the case
in real-world applications. This leads to decreased model transfer effects when
the new class distribution differs significantly from the learned classes.
Research into Cross-Domain Few-Shot (CDFS) has emerged to address this issue,
forming a more challenging and realistic setting. In this survey, we provide a
detailed taxonomy of CDFS from the problem setting and corresponding solutions
view. We summarise the existing CDFS network architectures and discuss the
solution ideas for each direction the taxonomy indicates. Furthermore, we
introduce various CDFS downstream applications and outline classification,
detection, and segmentation benchmarks and corresponding standards for
evaluation. We also discuss the challenges of CDFS research and explore
potential directions for future investigation. Through this review, we aim to
provide comprehensive guidance on CDFS research, enabling researchers to gain
insight into the state-of-the-art while allowing them to build upon existing
solutions to develop their own CDFS models.
| no_new_dataset | 0.949623 |
2303.09269 | Jes\'us Molina Rodr\'iguez-De-Vera | Pablo Villacorta, Jes\'us M. Rodr\'iguez-de-Vera, Marc Bola\~nos,
Ignacio Saras\'ua, Bhalaji Nagarajan, Petia Radeva | ELFIS: Expert Learning for Fine-grained Image Recognition Using Subsets | Pablo Villacorta and Jes\'us M. Rodr\'iguez-de-Vera contributed
equally to this work. 16 pages, 10 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Fine-Grained Visual Recognition (FGVR) tackles the problem of distinguishing
highly similar categories. One of the main approaches to FGVR, namely subset
learning, tries to leverage information from existing class taxonomies to
improve the performance of deep neural networks. However, these methods rely on
the existence of handcrafted hierarchies that are not necessarily optimal for
the models. In this paper, we propose ELFIS, an expert learning framework for
FGVR that clusters categories of the dataset into meta-categories using both
dataset-inherent lexical and model-specific information. A set of neural
networks-based experts are trained focusing on the meta-categories and are
integrated into a multi-task framework. Extensive experimentation shows
improvements in the SoTA FGVR benchmarks of up to +1.3% of accuracy using both
CNNs and transformer-based networks. Overall, the obtained results evidence
that ELFIS can be applied on top of any classification model, enabling the
obtention of SoTA results. The source code will be made public soon.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 12:45:19 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Villacorta",
"Pablo",
""
],
[
"Rodríguez-de-Vera",
"Jesús M.",
""
],
[
"Bolaños",
"Marc",
""
],
[
"Sarasúa",
"Ignacio",
""
],
[
"Nagarajan",
"Bhalaji",
""
],
[
"Radeva",
"Petia",
""
]
] | TITLE: ELFIS: Expert Learning for Fine-grained Image Recognition Using Subsets
ABSTRACT: Fine-Grained Visual Recognition (FGVR) tackles the problem of distinguishing
highly similar categories. One of the main approaches to FGVR, namely subset
learning, tries to leverage information from existing class taxonomies to
improve the performance of deep neural networks. However, these methods rely on
the existence of handcrafted hierarchies that are not necessarily optimal for
the models. In this paper, we propose ELFIS, an expert learning framework for
FGVR that clusters categories of the dataset into meta-categories using both
dataset-inherent lexical and model-specific information. A set of neural
networks-based experts are trained focusing on the meta-categories and are
integrated into a multi-task framework. Extensive experimentation shows
improvements in the SoTA FGVR benchmarks of up to +1.3% of accuracy using both
CNNs and transformer-based networks. Overall, the obtained results evidence
that ELFIS can be applied on top of any classification model, enabling the
obtention of SoTA results. The source code will be made public soon.
| no_new_dataset | 0.944944 |
2303.09273 | Yongchao Ye | Ying Wu, Yongchao Ye, Adnan Zeb, James J.Q. Yu, Zheng Wang | Adaptive Modeling of Uncertainties for Traffic Forecasting | 14 pages, 5 figures | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep neural networks (DNNs) have emerged as a dominant approach for
developing traffic forecasting models. These models are typically trained to
minimize error on averaged test cases and produce a single-point prediction,
such as a scalar value for traffic speed or travel time. However, single-point
predictions fail to account for prediction uncertainty that is critical for
many transportation management scenarios, such as determining the best- or
worst-case arrival time. We present QuanTraffic, a generic framework to enhance
the capability of an arbitrary DNN model for uncertainty modeling. QuanTraffic
requires little human involvement and does not change the base DNN architecture
during deployment. Instead, it automatically learns a standard quantile
function during the DNN model training to produce a prediction interval for the
single-point prediction. The prediction interval defines a range where the true
value of the traffic prediction is likely to fall. Furthermore, QuanTraffic
develops an adaptive scheme that dynamically adjusts the prediction interval
based on the location and prediction window of the test input. We evaluated
QuanTraffic by applying it to five representative DNN models for traffic
forecasting across seven public datasets. We then compared QuanTraffic against
five uncertainty quantification methods. Compared to the baseline uncertainty
modeling techniques, QuanTraffic with base DNN architectures delivers
consistently better and more robust performance than the existing ones on the
reported datasets.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 12:56:13 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Wu",
"Ying",
""
],
[
"Ye",
"Yongchao",
""
],
[
"Zeb",
"Adnan",
""
],
[
"Yu",
"James J. Q.",
""
],
[
"Wang",
"Zheng",
""
]
] | TITLE: Adaptive Modeling of Uncertainties for Traffic Forecasting
ABSTRACT: Deep neural networks (DNNs) have emerged as a dominant approach for
developing traffic forecasting models. These models are typically trained to
minimize error on averaged test cases and produce a single-point prediction,
such as a scalar value for traffic speed or travel time. However, single-point
predictions fail to account for prediction uncertainty that is critical for
many transportation management scenarios, such as determining the best- or
worst-case arrival time. We present QuanTraffic, a generic framework to enhance
the capability of an arbitrary DNN model for uncertainty modeling. QuanTraffic
requires little human involvement and does not change the base DNN architecture
during deployment. Instead, it automatically learns a standard quantile
function during the DNN model training to produce a prediction interval for the
single-point prediction. The prediction interval defines a range where the true
value of the traffic prediction is likely to fall. Furthermore, QuanTraffic
develops an adaptive scheme that dynamically adjusts the prediction interval
based on the location and prediction window of the test input. We evaluated
QuanTraffic by applying it to five representative DNN models for traffic
forecasting across seven public datasets. We then compared QuanTraffic against
five uncertainty quantification methods. Compared to the baseline uncertainty
modeling techniques, QuanTraffic with base DNN architectures delivers
consistently better and more robust performance than the existing ones on the
reported datasets.
| no_new_dataset | 0.947527 |
2303.09278 | Songjun Cao | Yanzhe Fu, Yueteng Kang, Songjun Cao, Long Ma | DistillW2V2: A Small and Streaming Wav2vec 2.0 Based ASR Model | null | null | null | null | eess.AS cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Wav2vec 2.0 (W2V2) has shown impressive performance in automatic speech
recognition (ASR). However, the large model size and the non-streaming
architecture make it hard to be used under low-resource or streaming scenarios.
In this work, we propose a two-stage knowledge distillation method to solve
these two problems: the first step is to make the big and non-streaming teacher
model smaller, and the second step is to make it streaming. Specially, we adopt
the MSE loss for the distillation of hidden layers and the modified LF-MMI loss
for the distillation of the prediction layer. Experiments are conducted on
Gigaspeech, Librispeech, and an in-house dataset. The results show that the
distilled student model (DistillW2V2) we finally get is 8x faster and 12x
smaller than the original teacher model. For the 480ms latency setup, the
DistillW2V2's relative word error rate (WER) degradation varies from 9% to
23.4% on test sets, which reveals a promising way to extend the W2V2's
application scope.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 12:59:17 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Fu",
"Yanzhe",
""
],
[
"Kang",
"Yueteng",
""
],
[
"Cao",
"Songjun",
""
],
[
"Ma",
"Long",
""
]
] | TITLE: DistillW2V2: A Small and Streaming Wav2vec 2.0 Based ASR Model
ABSTRACT: Wav2vec 2.0 (W2V2) has shown impressive performance in automatic speech
recognition (ASR). However, the large model size and the non-streaming
architecture make it hard to be used under low-resource or streaming scenarios.
In this work, we propose a two-stage knowledge distillation method to solve
these two problems: the first step is to make the big and non-streaming teacher
model smaller, and the second step is to make it streaming. Specially, we adopt
the MSE loss for the distillation of hidden layers and the modified LF-MMI loss
for the distillation of the prediction layer. Experiments are conducted on
Gigaspeech, Librispeech, and an in-house dataset. The results show that the
distilled student model (DistillW2V2) we finally get is 8x faster and 12x
smaller than the original teacher model. For the 480ms latency setup, the
DistillW2V2's relative word error rate (WER) degradation varies from 9% to
23.4% on test sets, which reveals a promising way to extend the W2V2's
application scope.
| no_new_dataset | 0.944995 |
2303.09283 | Rafael Rosales | Rafael Rosales, Pablo Munoz, Michael Paulitsch | Exploring Resiliency to Natural Image Corruptions in Deep Learning using
Design Diversity | null | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | In this paper, we investigate the relationship between diversity metrics,
accuracy, and resiliency to natural image corruptions of Deep Learning (DL)
image classifier ensembles. We investigate the potential of an
attribution-based diversity metric to improve the known accuracy-diversity
trade-off of the typical prediction-based diversity. Our motivation is based on
analytical studies of design diversity that have shown that a reduction of
common failure modes is possible if diversity of design choices is achieved.
Using ResNet50 as a comparison baseline, we evaluate the resiliency of
multiple individual DL model architectures against dataset distribution shifts
corresponding to natural image corruptions. We compare ensembles created with
diverse model architectures trained either independently or through a Neural
Architecture Search technique and evaluate the correlation of prediction-based
and attribution-based diversity to the final ensemble accuracy. We evaluate a
set of diversity enforcement heuristics based on negative correlation learning
to assess the final ensemble resilience to natural image corruptions and
inspect the resulting prediction, activation, and attribution diversity.
Our key observations are: 1) model architecture is more important for
resiliency than model size or model accuracy, 2) attribution-based diversity is
less negatively correlated to the ensemble accuracy than prediction-based
diversity, 3) a balanced loss function of individual and ensemble accuracy
creates more resilient ensembles for image natural corruptions, 4) architecture
diversity produces more diversity in all explored diversity metrics:
predictions, attributions, and activations.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 08:54:10 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Rosales",
"Rafael",
""
],
[
"Munoz",
"Pablo",
""
],
[
"Paulitsch",
"Michael",
""
]
] | TITLE: Exploring Resiliency to Natural Image Corruptions in Deep Learning using
Design Diversity
ABSTRACT: In this paper, we investigate the relationship between diversity metrics,
accuracy, and resiliency to natural image corruptions of Deep Learning (DL)
image classifier ensembles. We investigate the potential of an
attribution-based diversity metric to improve the known accuracy-diversity
trade-off of the typical prediction-based diversity. Our motivation is based on
analytical studies of design diversity that have shown that a reduction of
common failure modes is possible if diversity of design choices is achieved.
Using ResNet50 as a comparison baseline, we evaluate the resiliency of
multiple individual DL model architectures against dataset distribution shifts
corresponding to natural image corruptions. We compare ensembles created with
diverse model architectures trained either independently or through a Neural
Architecture Search technique and evaluate the correlation of prediction-based
and attribution-based diversity to the final ensemble accuracy. We evaluate a
set of diversity enforcement heuristics based on negative correlation learning
to assess the final ensemble resilience to natural image corruptions and
inspect the resulting prediction, activation, and attribution diversity.
Our key observations are: 1) model architecture is more important for
resiliency than model size or model accuracy, 2) attribution-based diversity is
less negatively correlated to the ensemble accuracy than prediction-based
diversity, 3) a balanced loss function of individual and ensemble accuracy
creates more resilient ensembles for image natural corruptions, 4) architecture
diversity produces more diversity in all explored diversity metrics:
predictions, attributions, and activations.
| no_new_dataset | 0.955026 |
2303.09295 | Zhendong Wang | Zhendong Wang, Jianmin Bao, Wengang Zhou, Weilun Wang, Hezhen Hu, Hong
Chen, Houqiang Li | DIRE for Diffusion-Generated Image Detection | A general diffusion-generated image detector | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Diffusion models have shown remarkable success in visual synthesis, but have
also raised concerns about potential abuse for malicious purposes. In this
paper, we seek to build a detector for telling apart real images from
diffusion-generated images. We find that existing detectors struggle to detect
images generated by diffusion models, even if we include generated images from
a specific diffusion model in their training data. To address this issue, we
propose a novel image representation called DIffusion Reconstruction Error
(DIRE), which measures the error between an input image and its reconstruction
counterpart by a pre-trained diffusion model. We observe that
diffusion-generated images can be approximately reconstructed by a diffusion
model while real images cannot. It provides a hint that DIRE can serve as a
bridge to distinguish generated and real images. DIRE provides an effective way
to detect images generated by most diffusion models, and it is general for
detecting generated images from unseen diffusion models and robust to various
perturbations. Furthermore, we establish a comprehensive diffusion-generated
benchmark including images generated by eight diffusion models to evaluate the
performance of diffusion-generated image detectors. Extensive experiments on
our collected benchmark demonstrate that DIRE exhibits superiority over
previous generated-image detectors. The code and dataset are available at
https://github.com/ZhendongWang6/DIRE.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 13:15:03 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Wang",
"Zhendong",
""
],
[
"Bao",
"Jianmin",
""
],
[
"Zhou",
"Wengang",
""
],
[
"Wang",
"Weilun",
""
],
[
"Hu",
"Hezhen",
""
],
[
"Chen",
"Hong",
""
],
[
"Li",
"Houqiang",
""
]
] | TITLE: DIRE for Diffusion-Generated Image Detection
ABSTRACT: Diffusion models have shown remarkable success in visual synthesis, but have
also raised concerns about potential abuse for malicious purposes. In this
paper, we seek to build a detector for telling apart real images from
diffusion-generated images. We find that existing detectors struggle to detect
images generated by diffusion models, even if we include generated images from
a specific diffusion model in their training data. To address this issue, we
propose a novel image representation called DIffusion Reconstruction Error
(DIRE), which measures the error between an input image and its reconstruction
counterpart by a pre-trained diffusion model. We observe that
diffusion-generated images can be approximately reconstructed by a diffusion
model while real images cannot. It provides a hint that DIRE can serve as a
bridge to distinguish generated and real images. DIRE provides an effective way
to detect images generated by most diffusion models, and it is general for
detecting generated images from unseen diffusion models and robust to various
perturbations. Furthermore, we establish a comprehensive diffusion-generated
benchmark including images generated by eight diffusion models to evaluate the
performance of diffusion-generated image detectors. Extensive experiments on
our collected benchmark demonstrate that DIRE exhibits superiority over
previous generated-image detectors. The code and dataset are available at
https://github.com/ZhendongWang6/DIRE.
| new_dataset | 0.91848 |
2303.09305 | Jing Mai | Jing Mai and Jiarui Wang and Zhixiong Di and Yibo Lin | Multi-Electrostatic FPGA Placement Considering SLICEL-SLICEM
Heterogeneity, Clock Feasibility, and Timing Optimization | null | null | null | null | cs.AR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | When modern FPGA architecture becomes increasingly complicated, modern FPGA
placement is a mixed optimization problem with multiple objectives, including
wirelength, routability, timing closure, and clock feasibility. Typical FPGA
devices nowadays consist of heterogeneous SLICEs like SLICEL and SLICEM. The
resources of a SLICE can be configured to {LUT, FF, distributed RAM, SHIFT,
CARRY}. Besides such heterogeneity, advanced FPGA architectures also bring
complicated constraints like timing, clock routing, carry chain alignment, etc.
The above heterogeneity and constraints impose increasing challenges to FPGA
placement algorithms.
In this work, we propose a multi-electrostatic FPGA placer considering the
aforementioned SLICEL-SLICEM heterogeneity under timing, clock routing and
carry chain alignment constraints. We first propose an effective SLICEL-SLICEM
heterogeneity model with a novel electrostatic-based density formulation. We
also design a dynamically adjusted preconditioning and carry chain alignment
technique to stabilize the optimization convergence. We then propose a
timing-driven net weighting scheme to incorporate timing optimization. Finally,
we put forward a nested Lagrangian relaxation-based placement framework to
incorporate the optimization objectives of wirelength, routability, timing, and
clock feasibility. Experimental results on both academic and industrial
benchmarks demonstrate that our placer outperforms the state-of-the-art placers
in quality and efficiency.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 13:30:23 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Mai",
"Jing",
""
],
[
"Wang",
"Jiarui",
""
],
[
"Di",
"Zhixiong",
""
],
[
"Lin",
"Yibo",
""
]
] | TITLE: Multi-Electrostatic FPGA Placement Considering SLICEL-SLICEM
Heterogeneity, Clock Feasibility, and Timing Optimization
ABSTRACT: When modern FPGA architecture becomes increasingly complicated, modern FPGA
placement is a mixed optimization problem with multiple objectives, including
wirelength, routability, timing closure, and clock feasibility. Typical FPGA
devices nowadays consist of heterogeneous SLICEs like SLICEL and SLICEM. The
resources of a SLICE can be configured to {LUT, FF, distributed RAM, SHIFT,
CARRY}. Besides such heterogeneity, advanced FPGA architectures also bring
complicated constraints like timing, clock routing, carry chain alignment, etc.
The above heterogeneity and constraints impose increasing challenges to FPGA
placement algorithms.
In this work, we propose a multi-electrostatic FPGA placer considering the
aforementioned SLICEL-SLICEM heterogeneity under timing, clock routing and
carry chain alignment constraints. We first propose an effective SLICEL-SLICEM
heterogeneity model with a novel electrostatic-based density formulation. We
also design a dynamically adjusted preconditioning and carry chain alignment
technique to stabilize the optimization convergence. We then propose a
timing-driven net weighting scheme to incorporate timing optimization. Finally,
we put forward a nested Lagrangian relaxation-based placement framework to
incorporate the optimization objectives of wirelength, routability, timing, and
clock feasibility. Experimental results on both academic and industrial
benchmarks demonstrate that our placer outperforms the state-of-the-art placers
in quality and efficiency.
| no_new_dataset | 0.94743 |
2303.09310 | Bo Dang | Yansheng Li, Bo Dang, Wanchun Li, Yongjun Zhang | GLH-Water: A Large-Scale Dataset for Global Surface Water Detection in
Large-Size Very-High-Resolution Satellite Imagery | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Global surface water detection in very-high-resolution (VHR) satellite
imagery can directly serve major applications such as refined flood mapping and
water resource assessment. Although achievements have been made in detecting
surface water in small-size satellite images corresponding to local geographic
scales, datasets and methods suitable for mapping and analyzing global surface
water have yet to be explored. To encourage the development of this task and
facilitate the implementation of relevant applications, we propose the
GLH-water dataset that consists of 250 satellite images and manually labeled
surface water annotations that are distributed globally and contain water
bodies exhibiting a wide variety of types (e.g., rivers, lakes, and ponds in
forests, irrigated fields, bare areas, and urban areas). Each image is of the
size 12,800 $\times$ 12,800 pixels at 0.3 meter spatial resolution. To build a
benchmark for GLH-water, we perform extensive experiments employing
representative surface water detection models, popular semantic segmentation
models, and ultra-high resolution segmentation models. Furthermore, we also
design a strong baseline with the novel pyramid consistency loss (PCL) to
initially explore this challenge. Finally, we implement the cross-dataset and
pilot area generalization experiments, and the superior performance illustrates
the strong generalization and practical application of GLH-water. The dataset
is available at https://jack-bo1220.github.io/project/GLH-water.html.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 13:35:56 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Li",
"Yansheng",
""
],
[
"Dang",
"Bo",
""
],
[
"Li",
"Wanchun",
""
],
[
"Zhang",
"Yongjun",
""
]
] | TITLE: GLH-Water: A Large-Scale Dataset for Global Surface Water Detection in
Large-Size Very-High-Resolution Satellite Imagery
ABSTRACT: Global surface water detection in very-high-resolution (VHR) satellite
imagery can directly serve major applications such as refined flood mapping and
water resource assessment. Although achievements have been made in detecting
surface water in small-size satellite images corresponding to local geographic
scales, datasets and methods suitable for mapping and analyzing global surface
water have yet to be explored. To encourage the development of this task and
facilitate the implementation of relevant applications, we propose the
GLH-water dataset that consists of 250 satellite images and manually labeled
surface water annotations that are distributed globally and contain water
bodies exhibiting a wide variety of types (e.g., rivers, lakes, and ponds in
forests, irrigated fields, bare areas, and urban areas). Each image is of the
size 12,800 $\times$ 12,800 pixels at 0.3 meter spatial resolution. To build a
benchmark for GLH-water, we perform extensive experiments employing
representative surface water detection models, popular semantic segmentation
models, and ultra-high resolution segmentation models. Furthermore, we also
design a strong baseline with the novel pyramid consistency loss (PCL) to
initially explore this challenge. Finally, we implement the cross-dataset and
pilot area generalization experiments, and the superior performance illustrates
the strong generalization and practical application of GLH-water. The dataset
is available at https://jack-bo1220.github.io/project/GLH-water.html.
| new_dataset | 0.967163 |
2303.09322 | Anthony Onoja | Anthony Onoja, Francesco Raimondi | Interpretability from a new lens: Integrating Stratification and Domain
knowledge for Biomedical Applications | 10 pages concept paper | null | null | null | cs.LG cs.HC q-bio.QM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The use of machine learning (ML) techniques in the biomedical field has
become increasingly important, particularly with the large amounts of data
generated by the aftermath of the COVID-19 pandemic. However, due to the
complex nature of biomedical datasets and the use of black-box ML models, a
lack of trust and adoption by domain experts can arise. In response,
interpretable ML (IML) approaches have been developed, but the curse of
dimensionality in biomedical datasets can lead to model instability. This paper
proposes a novel computational strategy for the stratification of biomedical
problem datasets into k-fold cross-validation (CVs) and integrating domain
knowledge interpretation techniques embedded into the current state-of-the-art
IML frameworks. This approach can improve model stability, establish trust, and
provide explanations for outcomes generated by trained IML models.
Specifically, the model outcome, such as aggregated feature weight importance,
can be linked to further domain knowledge interpretations using techniques like
pathway functional enrichment, drug targeting, and repurposing databases.
Additionally, involving end-users and clinicians in focus group discussions
before and after the choice of IML framework can help guide testable
hypotheses, improve performance metrics, and build trustworthy and usable IML
solutions in the biomedical field. Overall, this study highlights the potential
of combining advanced computational techniques with domain knowledge
interpretation to enhance the effectiveness of IML solutions in the context of
complex biomedical datasets.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 12:02:02 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Onoja",
"Anthony",
""
],
[
"Raimondi",
"Francesco",
""
]
] | TITLE: Interpretability from a new lens: Integrating Stratification and Domain
knowledge for Biomedical Applications
ABSTRACT: The use of machine learning (ML) techniques in the biomedical field has
become increasingly important, particularly with the large amounts of data
generated by the aftermath of the COVID-19 pandemic. However, due to the
complex nature of biomedical datasets and the use of black-box ML models, a
lack of trust and adoption by domain experts can arise. In response,
interpretable ML (IML) approaches have been developed, but the curse of
dimensionality in biomedical datasets can lead to model instability. This paper
proposes a novel computational strategy for the stratification of biomedical
problem datasets into k-fold cross-validation (CVs) and integrating domain
knowledge interpretation techniques embedded into the current state-of-the-art
IML frameworks. This approach can improve model stability, establish trust, and
provide explanations for outcomes generated by trained IML models.
Specifically, the model outcome, such as aggregated feature weight importance,
can be linked to further domain knowledge interpretations using techniques like
pathway functional enrichment, drug targeting, and repurposing databases.
Additionally, involving end-users and clinicians in focus group discussions
before and after the choice of IML framework can help guide testable
hypotheses, improve performance metrics, and build trustworthy and usable IML
solutions in the biomedical field. Overall, this study highlights the potential
of combining advanced computational techniques with domain knowledge
interpretation to enhance the effectiveness of IML solutions in the context of
complex biomedical datasets.
| no_new_dataset | 0.94545 |
2303.09336 | Lin Xi | Lin Xi, Weihai Chen, Changchen Zhao, Xingming Wu, and Jianhua Wang | Image Enhancement for Remote Photoplethysmography in a Low-Light
Environment | Accepted by FG2020 | null | 10.1109/FG47880.2020.00076 | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | With the improvement of sensor technology and significant algorithmic
advances, the accuracy of remote heart rate monitoring technology has been
significantly improved. Despite of the significant algorithmic advances, the
performance of rPPG algorithm can degrade in the long-term, high-intensity
continuous work occurred in evenings or insufficient light environments. One of
the main challenges is that the lost facial details and low contrast cause the
failure of detection and tracking. Also, insufficient lighting in video
capturing hurts the quality of physiological signal. In this paper, we collect
a large-scale dataset that was designed for remote heart rate estimation
recorded with various illumination variations to evaluate the performance of
the rPPG algorithm (Green, ICA, and POS). We also propose a low-light
enhancement solution (technical solution) for remote heart rate estimation
under the low-light condition. Using collected dataset, we found 1) face
detection algorithm cannot detect faces in video captured in low light
conditions; 2) A decrease in the amplitude of the pulsatile signal will lead to
the noise signal to be in the dominant position; and 3) the chrominance-based
method suffers from the limitation in the assumption about skin-tone will not
hold, and Green and ICA method receive less influence than POS in dark
illuminance environment. The proposed solution for rPPG process is effective to
detect and improve the signal-to-noise ratio and precision of the pulsatile
signal.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 14:18:48 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Xi",
"Lin",
""
],
[
"Chen",
"Weihai",
""
],
[
"Zhao",
"Changchen",
""
],
[
"Wu",
"Xingming",
""
],
[
"Wang",
"Jianhua",
""
]
] | TITLE: Image Enhancement for Remote Photoplethysmography in a Low-Light
Environment
ABSTRACT: With the improvement of sensor technology and significant algorithmic
advances, the accuracy of remote heart rate monitoring technology has been
significantly improved. Despite of the significant algorithmic advances, the
performance of rPPG algorithm can degrade in the long-term, high-intensity
continuous work occurred in evenings or insufficient light environments. One of
the main challenges is that the lost facial details and low contrast cause the
failure of detection and tracking. Also, insufficient lighting in video
capturing hurts the quality of physiological signal. In this paper, we collect
a large-scale dataset that was designed for remote heart rate estimation
recorded with various illumination variations to evaluate the performance of
the rPPG algorithm (Green, ICA, and POS). We also propose a low-light
enhancement solution (technical solution) for remote heart rate estimation
under the low-light condition. Using collected dataset, we found 1) face
detection algorithm cannot detect faces in video captured in low light
conditions; 2) A decrease in the amplitude of the pulsatile signal will lead to
the noise signal to be in the dominant position; and 3) the chrominance-based
method suffers from the limitation in the assumption about skin-tone will not
hold, and Green and ICA method receive less influence than POS in dark
illuminance environment. The proposed solution for rPPG process is effective to
detect and improve the signal-to-noise ratio and precision of the pulsatile
signal.
| new_dataset | 0.96741 |
2303.09364 | Guru Ravi Shanker Ramaguru | R Guru Ravi Shanker, B Manikanta Gupta, BV Koushik, Vinoo Alluri | Tollywood Emotions: Annotation of Valence-Arousal in Telugu Song Lyrics | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Emotion recognition from a given music track has heavily relied on acoustic
features, social tags, and metadata but is seldom focused on lyrics. There are
no datasets of Indian language songs that contain both valence and arousal
manual ratings of lyrics. We present a new manually annotated dataset of Telugu
songs' lyrics collected from Spotify with valence and arousal annotated on a
discrete scale. A fairly high inter-annotator agreement was observed for both
valence and arousal. Subsequently, we create two music emotion recognition
models by using two classification techniques to identify valence, arousal and
respective emotion quadrant from lyrics. Support vector machine (SVM) with term
frequency-inverse document frequency (TF-IDF) features and fine-tuning the
pre-trained XLMRoBERTa (XLM-R) model were used for valence, arousal and
quadrant classification tasks. Fine-tuned XLMRoBERTa performs better than the
SVM by improving macro-averaged F1-scores of 54.69%, 67.61%, 34.13% to 77.90%,
80.71% and 58.33% for valence, arousal and quadrant classifications,
respectively, on 10-fold cross-validation. In addition, we compare our lyrics
annotations with Spotify's annotations of valence and energy (same as arousal),
which are based on entire music tracks. The implications of our findings are
discussed. Finally, we make the dataset publicly available with lyrics,
annotations and Spotify IDs.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 14:47:52 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Shanker",
"R Guru Ravi",
""
],
[
"Gupta",
"B Manikanta",
""
],
[
"Koushik",
"BV",
""
],
[
"Alluri",
"Vinoo",
""
]
] | TITLE: Tollywood Emotions: Annotation of Valence-Arousal in Telugu Song Lyrics
ABSTRACT: Emotion recognition from a given music track has heavily relied on acoustic
features, social tags, and metadata but is seldom focused on lyrics. There are
no datasets of Indian language songs that contain both valence and arousal
manual ratings of lyrics. We present a new manually annotated dataset of Telugu
songs' lyrics collected from Spotify with valence and arousal annotated on a
discrete scale. A fairly high inter-annotator agreement was observed for both
valence and arousal. Subsequently, we create two music emotion recognition
models by using two classification techniques to identify valence, arousal and
respective emotion quadrant from lyrics. Support vector machine (SVM) with term
frequency-inverse document frequency (TF-IDF) features and fine-tuning the
pre-trained XLMRoBERTa (XLM-R) model were used for valence, arousal and
quadrant classification tasks. Fine-tuned XLMRoBERTa performs better than the
SVM by improving macro-averaged F1-scores of 54.69%, 67.61%, 34.13% to 77.90%,
80.71% and 58.33% for valence, arousal and quadrant classifications,
respectively, on 10-fold cross-validation. In addition, we compare our lyrics
annotations with Spotify's annotations of valence and energy (same as arousal),
which are based on entire music tracks. The implications of our findings are
discussed. Finally, we make the dataset publicly available with lyrics,
annotations and Spotify IDs.
| new_dataset | 0.959649 |
2303.09370 | Zhihao Chen | Zhihao Chen, Liang Wan, Yefan Xiao, Lei Zhu, Huazhu Fu | Learning Physical-Spatio-Temporal Features for Video Shadow Removal | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Shadow removal in a single image has received increasing attention in recent
years. However, removing shadows over dynamic scenes remains largely
under-explored. In this paper, we propose the first data-driven video shadow
removal model, termed PSTNet, by exploiting three essential characteristics of
video shadows, i.e., physical property, spatio relation, and temporal
coherence. Specifically, a dedicated physical branch was established to conduct
local illumination estimation, which is more applicable for scenes with complex
lighting and textures, and then enhance the physical features via a mask-guided
attention strategy. Then, we develop a progressive aggregation module to
enhance the spatio and temporal characteristics of features maps, and
effectively integrate the three kinds of features. Furthermore, to tackle the
lack of datasets of paired shadow videos, we synthesize a dataset (SVSRD-85)
with aid of the popular game GTAV by controlling the switch of the shadow
renderer. Experiments against 9 state-of-the-art models, including image shadow
removers and image/video restoration methods, show that our method improves the
best SOTA in terms of RMSE error for the shadow area by 14.7. In addition, we
develop a lightweight model adaptation strategy to make our synthetic-driven
model effective in real world scenes. The visual comparison on the public
SBU-TimeLapse dataset verifies the generalization ability of our model in real
scenes.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 14:55:31 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Chen",
"Zhihao",
""
],
[
"Wan",
"Liang",
""
],
[
"Xiao",
"Yefan",
""
],
[
"Zhu",
"Lei",
""
],
[
"Fu",
"Huazhu",
""
]
] | TITLE: Learning Physical-Spatio-Temporal Features for Video Shadow Removal
ABSTRACT: Shadow removal in a single image has received increasing attention in recent
years. However, removing shadows over dynamic scenes remains largely
under-explored. In this paper, we propose the first data-driven video shadow
removal model, termed PSTNet, by exploiting three essential characteristics of
video shadows, i.e., physical property, spatio relation, and temporal
coherence. Specifically, a dedicated physical branch was established to conduct
local illumination estimation, which is more applicable for scenes with complex
lighting and textures, and then enhance the physical features via a mask-guided
attention strategy. Then, we develop a progressive aggregation module to
enhance the spatio and temporal characteristics of features maps, and
effectively integrate the three kinds of features. Furthermore, to tackle the
lack of datasets of paired shadow videos, we synthesize a dataset (SVSRD-85)
with aid of the popular game GTAV by controlling the switch of the shadow
renderer. Experiments against 9 state-of-the-art models, including image shadow
removers and image/video restoration methods, show that our method improves the
best SOTA in terms of RMSE error for the shadow area by 14.7. In addition, we
develop a lightweight model adaptation strategy to make our synthetic-driven
model effective in real world scenes. The visual comparison on the public
SBU-TimeLapse dataset verifies the generalization ability of our model in real
scenes.
| new_dataset | 0.964689 |
2303.09381 | Junchen Yang | Junchen Yang, Ofir Lindenbaum, Yuval Kluger, Ariel Jaffe | Multi-modal Differentiable Unsupervised Feature Selection | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Multi-modal high throughput biological data presents a great scientific
opportunity and a significant computational challenge. In multi-modal
measurements, every sample is observed simultaneously by two or more sets of
sensors. In such settings, many observed variables in both modalities are often
nuisance and do not carry information about the phenomenon of interest. Here,
we propose a multi-modal unsupervised feature selection framework: identifying
informative variables based on coupled high-dimensional measurements. Our
method is designed to identify features associated with two types of latent
low-dimensional structures: (i) shared structures that govern the observations
in both modalities and (ii) differential structures that appear in only one
modality. To that end, we propose two Laplacian-based scoring operators. We
incorporate the scores with differentiable gates that mask nuisance features
and enhance the accuracy of the structure captured by the graph Laplacian. The
performance of the new scheme is illustrated using synthetic and real datasets,
including an extended biological application to single-cell multi-omics.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 15:11:17 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Yang",
"Junchen",
""
],
[
"Lindenbaum",
"Ofir",
""
],
[
"Kluger",
"Yuval",
""
],
[
"Jaffe",
"Ariel",
""
]
] | TITLE: Multi-modal Differentiable Unsupervised Feature Selection
ABSTRACT: Multi-modal high throughput biological data presents a great scientific
opportunity and a significant computational challenge. In multi-modal
measurements, every sample is observed simultaneously by two or more sets of
sensors. In such settings, many observed variables in both modalities are often
nuisance and do not carry information about the phenomenon of interest. Here,
we propose a multi-modal unsupervised feature selection framework: identifying
informative variables based on coupled high-dimensional measurements. Our
method is designed to identify features associated with two types of latent
low-dimensional structures: (i) shared structures that govern the observations
in both modalities and (ii) differential structures that appear in only one
modality. To that end, we propose two Laplacian-based scoring operators. We
incorporate the scores with differentiable gates that mask nuisance features
and enhance the accuracy of the structure captured by the graph Laplacian. The
performance of the new scheme is illustrated using synthetic and real datasets,
including an extended biological application to single-cell multi-omics.
| no_new_dataset | 0.947721 |
2303.09384 | Nicolas E. Diaz Ferreyra PhD | Catherine Tony, Markus Mutas, Nicol\'as E. D\'iaz Ferreyra and
Riccardo Scandariato | LLMSecEval: A Dataset of Natural Language Prompts for Security
Evaluations | Accepted at MSR '23 Data and Tool Showcase Track | null | null | null | cs.SE cs.IR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) like Codex are powerful tools for performing
code completion and code generation tasks as they are trained on billions of
lines of code from publicly available sources. Moreover, these models are
capable of generating code snippets from Natural Language (NL) descriptions by
learning languages and programming practices from public GitHub repositories.
Although LLMs promise an effortless NL-driven deployment of software
applications, the security of the code they generate has not been extensively
investigated nor documented. In this work, we present LLMSecEval, a dataset
containing 150 NL prompts that can be leveraged for assessing the security
performance of such models. Such prompts are NL descriptions of code snippets
prone to various security vulnerabilities listed in MITRE's Top 25 Common
Weakness Enumeration (CWE) ranking. Each prompt in our dataset comes with a
secure implementation example to facilitate comparative evaluations against
code produced by LLMs. As a practical application, we show how LLMSecEval can
be used for evaluating the security of snippets automatically generated from NL
descriptions.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 15:13:58 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Tony",
"Catherine",
""
],
[
"Mutas",
"Markus",
""
],
[
"Ferreyra",
"Nicolás E. Díaz",
""
],
[
"Scandariato",
"Riccardo",
""
]
] | TITLE: LLMSecEval: A Dataset of Natural Language Prompts for Security
Evaluations
ABSTRACT: Large Language Models (LLMs) like Codex are powerful tools for performing
code completion and code generation tasks as they are trained on billions of
lines of code from publicly available sources. Moreover, these models are
capable of generating code snippets from Natural Language (NL) descriptions by
learning languages and programming practices from public GitHub repositories.
Although LLMs promise an effortless NL-driven deployment of software
applications, the security of the code they generate has not been extensively
investigated nor documented. In this work, we present LLMSecEval, a dataset
containing 150 NL prompts that can be leveraged for assessing the security
performance of such models. Such prompts are NL descriptions of code snippets
prone to various security vulnerabilities listed in MITRE's Top 25 Common
Weakness Enumeration (CWE) ranking. Each prompt in our dataset comes with a
secure implementation example to facilitate comparative evaluations against
code produced by LLMs. As a practical application, we show how LLMSecEval can
be used for evaluating the security of snippets automatically generated from NL
descriptions.
| new_dataset | 0.959154 |
2303.09390 | Quanquan Gu | Weitong Zhang and Jiafan He and Zhiyuan Fan and Quanquan Gu | On the Interplay Between Misspecification and Sub-optimality Gap in
Linear Contextual Bandits | 28 pages, 2 figures, 2 tables | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study linear contextual bandits in the misspecified setting, where the
expected reward function can be approximated by a linear function class up to a
bounded misspecification level $\zeta>0$. We propose an algorithm based on a
novel data selection scheme, which only selects the contextual vectors with
large uncertainty for online regression. We show that, when the
misspecification level $\zeta$ is dominated by $\tilde O (\Delta / \sqrt{d})$
with $\Delta$ being the minimal sub-optimality gap and $d$ being the dimension
of the contextual vectors, our algorithm enjoys the same gap-dependent regret
bound $\tilde O (d^2/\Delta)$ as in the well-specified setting up to
logarithmic factors. In addition, we show that an existing algorithm SupLinUCB
(Chu et al., 2011) can also achieve a gap-dependent constant regret bound
without the knowledge of sub-optimality gap $\Delta$. Together with a lower
bound adapted from Lattimore et al. (2020), our result suggests an interplay
between misspecification level and the sub-optimality gap: (1) the linear
contextual bandit model is efficiently learnable when $\zeta \leq \tilde
O(\Delta / \sqrt{d})$; and (2) it is not efficiently learnable when $\zeta \geq
\tilde \Omega({\Delta} / {\sqrt{d}})$. Experiments on both synthetic and
real-world datasets corroborate our theoretical results.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 15:24:29 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Zhang",
"Weitong",
""
],
[
"He",
"Jiafan",
""
],
[
"Fan",
"Zhiyuan",
""
],
[
"Gu",
"Quanquan",
""
]
] | TITLE: On the Interplay Between Misspecification and Sub-optimality Gap in
Linear Contextual Bandits
ABSTRACT: We study linear contextual bandits in the misspecified setting, where the
expected reward function can be approximated by a linear function class up to a
bounded misspecification level $\zeta>0$. We propose an algorithm based on a
novel data selection scheme, which only selects the contextual vectors with
large uncertainty for online regression. We show that, when the
misspecification level $\zeta$ is dominated by $\tilde O (\Delta / \sqrt{d})$
with $\Delta$ being the minimal sub-optimality gap and $d$ being the dimension
of the contextual vectors, our algorithm enjoys the same gap-dependent regret
bound $\tilde O (d^2/\Delta)$ as in the well-specified setting up to
logarithmic factors. In addition, we show that an existing algorithm SupLinUCB
(Chu et al., 2011) can also achieve a gap-dependent constant regret bound
without the knowledge of sub-optimality gap $\Delta$. Together with a lower
bound adapted from Lattimore et al. (2020), our result suggests an interplay
between misspecification level and the sub-optimality gap: (1) the linear
contextual bandit model is efficiently learnable when $\zeta \leq \tilde
O(\Delta / \sqrt{d})$; and (2) it is not efficiently learnable when $\zeta \geq
\tilde \Omega({\Delta} / {\sqrt{d}})$. Experiments on both synthetic and
real-world datasets corroborate our theoretical results.
| no_new_dataset | 0.946843 |
2303.09406 | Chang Liu | Chang Liu and Sandra Paterlini | Stock Price Prediction Using Temporal Graph Model with Value Chain Data | null | null | null | null | q-fin.ST cs.LG q-fin.CP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Stock price prediction is a crucial element in financial trading as it allows
traders to make informed decisions about buying, selling, and holding stocks.
Accurate predictions of future stock prices can help traders optimize their
trading strategies and maximize their profits. In this paper, we introduce a
neural network-based stock return prediction method, the Long Short-Term Memory
Graph Convolutional Neural Network (LSTM-GCN) model, which combines the Graph
Convolutional Network (GCN) and Long Short-Term Memory (LSTM) Cells.
Specifically, the GCN is used to capture complex topological structures and
spatial dependence from value chain data, while the LSTM captures temporal
dependence and dynamic changes in stock returns data. We evaluated the LSTM-GCN
model on two datasets consisting of constituents of Eurostoxx 600 and S&P 500.
Our experiments demonstrate that the LSTM-GCN model can capture additional
information from value chain data that are not fully reflected in price data,
and the predictions outperform baseline models on both datasets.
| [
{
"version": "v1",
"created": "Tue, 7 Mar 2023 17:24:04 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Liu",
"Chang",
""
],
[
"Paterlini",
"Sandra",
""
]
] | TITLE: Stock Price Prediction Using Temporal Graph Model with Value Chain Data
ABSTRACT: Stock price prediction is a crucial element in financial trading as it allows
traders to make informed decisions about buying, selling, and holding stocks.
Accurate predictions of future stock prices can help traders optimize their
trading strategies and maximize their profits. In this paper, we introduce a
neural network-based stock return prediction method, the Long Short-Term Memory
Graph Convolutional Neural Network (LSTM-GCN) model, which combines the Graph
Convolutional Network (GCN) and Long Short-Term Memory (LSTM) Cells.
Specifically, the GCN is used to capture complex topological structures and
spatial dependence from value chain data, while the LSTM captures temporal
dependence and dynamic changes in stock returns data. We evaluated the LSTM-GCN
model on two datasets consisting of constituents of Eurostoxx 600 and S&P 500.
Our experiments demonstrate that the LSTM-GCN model can capture additional
information from value chain data that are not fully reflected in price data,
and the predictions outperform baseline models on both datasets.
| no_new_dataset | 0.954816 |
2303.09410 | Haibiao Xuan | Haibiao Xuan, Xiongzheng Li, Jinsong Zhang, Hongwen Zhang, Yebin Liu
and Kun Li | Narrator: Towards Natural Control of Human-Scene Interaction Generation
via Relationship Reasoning | Project page: http://cic.tju.edu.cn/faculty/likun/projects/Narrator | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Naturally controllable human-scene interaction (HSI) generation has an
important role in various fields, such as VR/AR content creation and
human-centered AI. However, existing methods are unnatural and unintuitive in
their controllability, which heavily limits their application in practice.
Therefore, we focus on a challenging task of naturally and controllably
generating realistic and diverse HSIs from textual descriptions. From human
cognition, the ideal generative model should correctly reason about spatial
relationships and interactive actions. To that end, we propose Narrator, a
novel relationship reasoning-based generative approach using a conditional
variation autoencoder for naturally controllable generation given a 3D scene
and a textual description. Also, we model global and local spatial
relationships in a 3D scene and a textual description respectively based on the
scene graph, and introduce a partlevel action mechanism to represent
interactions as atomic body part states. In particular, benefiting from our
relationship reasoning, we further propose a simple yet effective multi-human
generation strategy, which is the first exploration for controllable
multi-human scene interaction generation. Our extensive experiments and
perceptual studies show that Narrator can controllably generate diverse
interactions and significantly outperform existing works. The code and dataset
will be available for research purposes.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 15:44:15 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Xuan",
"Haibiao",
""
],
[
"Li",
"Xiongzheng",
""
],
[
"Zhang",
"Jinsong",
""
],
[
"Zhang",
"Hongwen",
""
],
[
"Liu",
"Yebin",
""
],
[
"Li",
"Kun",
""
]
] | TITLE: Narrator: Towards Natural Control of Human-Scene Interaction Generation
via Relationship Reasoning
ABSTRACT: Naturally controllable human-scene interaction (HSI) generation has an
important role in various fields, such as VR/AR content creation and
human-centered AI. However, existing methods are unnatural and unintuitive in
their controllability, which heavily limits their application in practice.
Therefore, we focus on a challenging task of naturally and controllably
generating realistic and diverse HSIs from textual descriptions. From human
cognition, the ideal generative model should correctly reason about spatial
relationships and interactive actions. To that end, we propose Narrator, a
novel relationship reasoning-based generative approach using a conditional
variation autoencoder for naturally controllable generation given a 3D scene
and a textual description. Also, we model global and local spatial
relationships in a 3D scene and a textual description respectively based on the
scene graph, and introduce a partlevel action mechanism to represent
interactions as atomic body part states. In particular, benefiting from our
relationship reasoning, we further propose a simple yet effective multi-human
generation strategy, which is the first exploration for controllable
multi-human scene interaction generation. Our extensive experiments and
perceptual studies show that Narrator can controllably generate diverse
interactions and significantly outperform existing works. The code and dataset
will be available for research purposes.
| no_new_dataset | 0.93196 |
2303.09427 | Sergio Tascon-Morales | Sergio Tascon-Morales and Pablo M\'arquez-Neila and Raphael Sznitman | Logical Implications for Visual Question Answering Consistency | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Despite considerable recent progress in Visual Question Answering (VQA)
models, inconsistent or contradictory answers continue to cast doubt on their
true reasoning capabilities. However, most proposed methods use indirect
strategies or strong assumptions on pairs of questions and answers to enforce
model consistency. Instead, we propose a novel strategy intended to improve
model performance by directly reducing logical inconsistencies. To do this, we
introduce a new consistency loss term that can be used by a wide range of the
VQA models and which relies on knowing the logical relation between pairs of
questions and answers. While such information is typically not available in VQA
datasets, we propose to infer these logical relations using a dedicated
language model and use these in our proposed consistency loss function. We
conduct extensive experiments on the VQA Introspect and DME datasets and show
that our method brings improvements to state-of-the-art VQA models, while being
robust across different architectures and settings.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 16:00:18 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Tascon-Morales",
"Sergio",
""
],
[
"Márquez-Neila",
"Pablo",
""
],
[
"Sznitman",
"Raphael",
""
]
] | TITLE: Logical Implications for Visual Question Answering Consistency
ABSTRACT: Despite considerable recent progress in Visual Question Answering (VQA)
models, inconsistent or contradictory answers continue to cast doubt on their
true reasoning capabilities. However, most proposed methods use indirect
strategies or strong assumptions on pairs of questions and answers to enforce
model consistency. Instead, we propose a novel strategy intended to improve
model performance by directly reducing logical inconsistencies. To do this, we
introduce a new consistency loss term that can be used by a wide range of the
VQA models and which relies on knowing the logical relation between pairs of
questions and answers. While such information is typically not available in VQA
datasets, we propose to infer these logical relations using a dedicated
language model and use these in our proposed consistency loss function. We
conduct extensive experiments on the VQA Introspect and DME datasets and show
that our method brings improvements to state-of-the-art VQA models, while being
robust across different architectures and settings.
| no_new_dataset | 0.941975 |
2303.09453 | Joakim Linja | Joakim Linja and Joonas H\"am\"al\"ainen and Antti Pihlajam\"aki and
Paavo Nieminen and Sami Malola and Hannu H\"akkinen and Tommi K\"arkk\"ainen | Knowledge Discovery from Atomic Structures using Feature Importances | 32 pages | null | null | null | cond-mat.mtrl-sci cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Molecular-level understanding of the interactions between the constituents of
an atomic structure is essential for designing novel materials in various
applications. This need goes beyond the basic knowledge of the number and types
of atoms, their chemical composition, and the character of the chemical
interactions. The bigger picture takes place on the quantum level which can be
addressed by using the Density-functional theory (DFT). Use of DFT, however, is
a computationally taxing process, and its results do not readily provide easily
interpretable insight into the atomic interactions which would be useful
information in material design. An alternative way to address atomic
interactions is to use an interpretable machine learning approach, where a
predictive DFT surrogate is constructed and analyzed. The purpose of this paper
is to propose such a procedure using a modification of the recently published
interpretable distance-based regression method. Our tests with a representative
benchmark set of molecules and a complex hybrid nanoparticle confirm the
viability and usefulness of the proposed approach.
| [
{
"version": "v1",
"created": "Tue, 28 Feb 2023 11:54:10 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Linja",
"Joakim",
""
],
[
"Hämäläinen",
"Joonas",
""
],
[
"Pihlajamäki",
"Antti",
""
],
[
"Nieminen",
"Paavo",
""
],
[
"Malola",
"Sami",
""
],
[
"Häkkinen",
"Hannu",
""
],
[
"Kärkkäinen",
"Tommi",
""
]
] | TITLE: Knowledge Discovery from Atomic Structures using Feature Importances
ABSTRACT: Molecular-level understanding of the interactions between the constituents of
an atomic structure is essential for designing novel materials in various
applications. This need goes beyond the basic knowledge of the number and types
of atoms, their chemical composition, and the character of the chemical
interactions. The bigger picture takes place on the quantum level which can be
addressed by using the Density-functional theory (DFT). Use of DFT, however, is
a computationally taxing process, and its results do not readily provide easily
interpretable insight into the atomic interactions which would be useful
information in material design. An alternative way to address atomic
interactions is to use an interpretable machine learning approach, where a
predictive DFT surrogate is constructed and analyzed. The purpose of this paper
is to propose such a procedure using a modification of the recently published
interpretable distance-based regression method. Our tests with a representative
benchmark set of molecules and a complex hybrid nanoparticle confirm the
viability and usefulness of the proposed approach.
| no_new_dataset | 0.943867 |
2303.09489 | Michael Zhang | Michael Zhang, Khaled K. Saab, Michael Poli, Tri Dao, Karan Goel, and
Christopher R\'e | Effectively Modeling Time Series with Simple Discrete State Spaces | 45 pages, 8 figures, 20 tables, ICLR 2023 | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Time series modeling is a well-established problem, which often requires that
methods (1) expressively represent complicated dependencies, (2) forecast long
horizons, and (3) efficiently train over long sequences. State-space models
(SSMs) are classical models for time series, and prior works combine SSMs with
deep learning layers for efficient sequence modeling. However, we find
fundamental limitations with these prior approaches, proving their SSM
representations cannot express autoregressive time series processes. We thus
introduce SpaceTime, a new state-space time series architecture that improves
all three criteria. For expressivity, we propose a new SSM parameterization
based on the companion matrix -- a canonical representation for discrete-time
processes -- which enables SpaceTime's SSM layers to learn desirable
autoregressive processes. For long horizon forecasting, we introduce a
"closed-loop" variation of the companion SSM, which enables SpaceTime to
predict many future time-steps by generating its own layer-wise inputs. For
efficient training and inference, we introduce an algorithm that reduces the
memory and compute of a forward pass with the companion matrix. With sequence
length $\ell$ and state-space size $d$, we go from $\tilde{O}(d \ell)$
na\"ively to $\tilde{O}(d + \ell)$. In experiments, our contributions lead to
state-of-the-art results on extensive and diverse benchmarks, with best or
second-best AUROC on 6 / 7 ECG and speech time series classification, and best
MSE on 14 / 16 Informer forecasting tasks. Furthermore, we find SpaceTime (1)
fits AR($p$) processes that prior deep SSMs fail on, (2) forecasts notably more
accurately on longer horizons than prior state-of-the-art, and (3) speeds up
training on real-world ETTh1 data by 73% and 80% relative wall-clock time over
Transformers and LSTMs.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:08:21 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Zhang",
"Michael",
""
],
[
"Saab",
"Khaled K.",
""
],
[
"Poli",
"Michael",
""
],
[
"Dao",
"Tri",
""
],
[
"Goel",
"Karan",
""
],
[
"Ré",
"Christopher",
""
]
] | TITLE: Effectively Modeling Time Series with Simple Discrete State Spaces
ABSTRACT: Time series modeling is a well-established problem, which often requires that
methods (1) expressively represent complicated dependencies, (2) forecast long
horizons, and (3) efficiently train over long sequences. State-space models
(SSMs) are classical models for time series, and prior works combine SSMs with
deep learning layers for efficient sequence modeling. However, we find
fundamental limitations with these prior approaches, proving their SSM
representations cannot express autoregressive time series processes. We thus
introduce SpaceTime, a new state-space time series architecture that improves
all three criteria. For expressivity, we propose a new SSM parameterization
based on the companion matrix -- a canonical representation for discrete-time
processes -- which enables SpaceTime's SSM layers to learn desirable
autoregressive processes. For long horizon forecasting, we introduce a
"closed-loop" variation of the companion SSM, which enables SpaceTime to
predict many future time-steps by generating its own layer-wise inputs. For
efficient training and inference, we introduce an algorithm that reduces the
memory and compute of a forward pass with the companion matrix. With sequence
length $\ell$ and state-space size $d$, we go from $\tilde{O}(d \ell)$
na\"ively to $\tilde{O}(d + \ell)$. In experiments, our contributions lead to
state-of-the-art results on extensive and diverse benchmarks, with best or
second-best AUROC on 6 / 7 ECG and speech time series classification, and best
MSE on 14 / 16 Informer forecasting tasks. Furthermore, we find SpaceTime (1)
fits AR($p$) processes that prior deep SSMs fail on, (2) forecasts notably more
accurately on longer horizons than prior state-of-the-art, and (3) speeds up
training on real-world ETTh1 data by 73% and 80% relative wall-clock time over
Transformers and LSTMs.
| no_new_dataset | 0.951684 |
2303.09494 | Eddardaa Benloussaief | Eddardaa Ben Loussaief, Hatem Rashwan, Mohammed Ayad, Mohammed Zakaria
Hassan, and Domenec Puig | Knowledge Distillation for Adaptive MRI Prostate Segmentation Based on
Limit-Trained Multi-Teacher Models | null | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | With numerous medical tasks, the performance of deep models has recently
experienced considerable improvements. These models are often adept learners.
Yet, their intricate architectural design and high computational complexity
make deploying them in clinical settings challenging, particularly with devices
with limited resources. To deal with this issue, Knowledge Distillation (KD)
has been proposed as a compression method and an acceleration technology. KD is
an efficient learning strategy that can transfer knowledge from a burdensome
model (i.e., teacher model) to a lightweight model (i.e., student model). Hence
we can obtain a compact model with low parameters with preserving the teacher's
performance. Therefore, we develop a KD-based deep model for prostate MRI
segmentation in this work by combining features-based distillation with
Kullback-Leibler divergence, Lovasz, and Dice losses. We further demonstrate
its effectiveness by applying two compression procedures: 1) distilling
knowledge to a student model from a single well-trained teacher, and 2) since
most of the medical applications have a small dataset, we train multiple
teachers that each one trained with a small set of images to learn an adaptive
student model as close to the teachers as possible considering the desired
accuracy and fast inference time. Extensive experiments were conducted on a
public multi-site prostate tumor dataset, showing that the proposed adaptation
KD strategy improves the dice similarity score by 9%, outperforming all tested
well-established baseline models.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:15:08 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Loussaief",
"Eddardaa Ben",
""
],
[
"Rashwan",
"Hatem",
""
],
[
"Ayad",
"Mohammed",
""
],
[
"Hassan",
"Mohammed Zakaria",
""
],
[
"Puig",
"Domenec",
""
]
] | TITLE: Knowledge Distillation for Adaptive MRI Prostate Segmentation Based on
Limit-Trained Multi-Teacher Models
ABSTRACT: With numerous medical tasks, the performance of deep models has recently
experienced considerable improvements. These models are often adept learners.
Yet, their intricate architectural design and high computational complexity
make deploying them in clinical settings challenging, particularly with devices
with limited resources. To deal with this issue, Knowledge Distillation (KD)
has been proposed as a compression method and an acceleration technology. KD is
an efficient learning strategy that can transfer knowledge from a burdensome
model (i.e., teacher model) to a lightweight model (i.e., student model). Hence
we can obtain a compact model with low parameters with preserving the teacher's
performance. Therefore, we develop a KD-based deep model for prostate MRI
segmentation in this work by combining features-based distillation with
Kullback-Leibler divergence, Lovasz, and Dice losses. We further demonstrate
its effectiveness by applying two compression procedures: 1) distilling
knowledge to a student model from a single well-trained teacher, and 2) since
most of the medical applications have a small dataset, we train multiple
teachers that each one trained with a small set of images to learn an adaptive
student model as close to the teachers as possible considering the desired
accuracy and fast inference time. Extensive experiments were conducted on a
public multi-site prostate tumor dataset, showing that the proposed adaptation
KD strategy improves the dice similarity score by 9%, outperforming all tested
well-established baseline models.
| no_new_dataset | 0.946101 |
2303.09497 | Lujuan Dang | Ziyue Huang, Lujuan Dang, Yuqing Xie, Wentao Ma, and Badong Chen | Gate Recurrent Unit Network based on Hilbert-Schmidt Independence
Criterion for State-of-Health Estimation | null | null | null | null | cs.LG cs.HC cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | State-of-health (SOH) estimation is a key step in ensuring the safe and
reliable operation of batteries. Due to issues such as varying data
distribution and sequence length in different cycles, most existing methods
require health feature extraction technique, which can be time-consuming and
labor-intensive. GRU can well solve this problem due to the simple structure
and superior performance, receiving widespread attentions. However, redundant
information still exists within the network and impacts the accuracy of SOH
estimation. To address this issue, a new GRU network based on Hilbert-Schmidt
Independence Criterion (GRU-HSIC) is proposed. First, a zero masking network is
used to transform all battery data measured with varying lengths every cycle
into sequences of the same length, while still retaining information about the
original data size in each cycle. Second, the Hilbert-Schmidt Independence
Criterion (HSIC) bottleneck, which evolved from Information Bottleneck (IB)
theory, is extended to GRU to compress the information from hidden layers. To
evaluate the proposed method, we conducted experiments on datasets from the
Center for Advanced Life Cycle Engineering (CALCE) of the University of
Maryland and NASA Ames Prognostics Center of Excellence. Experimental results
demonstrate that our model achieves higher accuracy than other recurrent
models.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:16:21 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Huang",
"Ziyue",
""
],
[
"Dang",
"Lujuan",
""
],
[
"Xie",
"Yuqing",
""
],
[
"Ma",
"Wentao",
""
],
[
"Chen",
"Badong",
""
]
] | TITLE: Gate Recurrent Unit Network based on Hilbert-Schmidt Independence
Criterion for State-of-Health Estimation
ABSTRACT: State-of-health (SOH) estimation is a key step in ensuring the safe and
reliable operation of batteries. Due to issues such as varying data
distribution and sequence length in different cycles, most existing methods
require health feature extraction technique, which can be time-consuming and
labor-intensive. GRU can well solve this problem due to the simple structure
and superior performance, receiving widespread attentions. However, redundant
information still exists within the network and impacts the accuracy of SOH
estimation. To address this issue, a new GRU network based on Hilbert-Schmidt
Independence Criterion (GRU-HSIC) is proposed. First, a zero masking network is
used to transform all battery data measured with varying lengths every cycle
into sequences of the same length, while still retaining information about the
original data size in each cycle. Second, the Hilbert-Schmidt Independence
Criterion (HSIC) bottleneck, which evolved from Information Bottleneck (IB)
theory, is extended to GRU to compress the information from hidden layers. To
evaluate the proposed method, we conducted experiments on datasets from the
Center for Advanced Life Cycle Engineering (CALCE) of the University of
Maryland and NASA Ames Prognostics Center of Excellence. Experimental results
demonstrate that our model achieves higher accuracy than other recurrent
models.
| no_new_dataset | 0.953144 |
2303.09527 | Zhenhuan Yang | Zhenhuan Yang, Yingqiang Ge, Congzhe Su, Dingxian Wang, Xiaoting Zhao,
Yiming Ying | Fairness-aware Differentially Private Collaborative Filtering | null | null | 10.1145/3543873.3587577 | null | cs.IR cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Recently, there has been an increasing adoption of differential privacy
guided algorithms for privacy-preserving machine learning tasks. However, the
use of such algorithms comes with trade-offs in terms of algorithmic fairness,
which has been widely acknowledged. Specifically, we have empirically observed
that the classical collaborative filtering method, trained by differentially
private stochastic gradient descent (DP-SGD), results in a disparate impact on
user groups with respect to different user engagement levels. This, in turn,
causes the original unfair model to become even more biased against inactive
users. To address the above issues, we propose \textbf{DP-Fair}, a two-stage
framework for collaborative filtering based algorithms. Specifically, it
combines differential privacy mechanisms with fairness constraints to protect
user privacy while ensuring fair recommendations. The experimental results,
based on Amazon datasets, and user history logs collected from Etsy, one of the
largest e-commerce platforms, demonstrate that our proposed method exhibits
superior performance in terms of both overall accuracy and user group fairness
on both shallow and deep recommendation models compared to vanilla DP-SGD.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:44:39 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Yang",
"Zhenhuan",
""
],
[
"Ge",
"Yingqiang",
""
],
[
"Su",
"Congzhe",
""
],
[
"Wang",
"Dingxian",
""
],
[
"Zhao",
"Xiaoting",
""
],
[
"Ying",
"Yiming",
""
]
] | TITLE: Fairness-aware Differentially Private Collaborative Filtering
ABSTRACT: Recently, there has been an increasing adoption of differential privacy
guided algorithms for privacy-preserving machine learning tasks. However, the
use of such algorithms comes with trade-offs in terms of algorithmic fairness,
which has been widely acknowledged. Specifically, we have empirically observed
that the classical collaborative filtering method, trained by differentially
private stochastic gradient descent (DP-SGD), results in a disparate impact on
user groups with respect to different user engagement levels. This, in turn,
causes the original unfair model to become even more biased against inactive
users. To address the above issues, we propose \textbf{DP-Fair}, a two-stage
framework for collaborative filtering based algorithms. Specifically, it
combines differential privacy mechanisms with fairness constraints to protect
user privacy while ensuring fair recommendations. The experimental results,
based on Amazon datasets, and user history logs collected from Etsy, one of the
largest e-commerce platforms, demonstrate that our proposed method exhibits
superior performance in terms of both overall accuracy and user group fairness
on both shallow and deep recommendation models compared to vanilla DP-SGD.
| no_new_dataset | 0.946597 |
2303.09528 | Shibashis Guha | Amin Falah, Shibashis Guha, Ashutosh Trivedi | Reinforcement Learning for Omega-Regular Specifications on
Continuous-Time MDP | Full version of paper accepted to ICAPS 2023 | null | null | null | cs.LG cs.AI math.OC | http://creativecommons.org/licenses/by/4.0/ | Continuous-time Markov decision processes (CTMDPs) are canonical models to
express sequential decision-making under dense-time and stochastic
environments. When the stochastic evolution of the environment is only
available via sampling, model-free reinforcement learning (RL) is the
algorithm-of-choice to compute optimal decision sequence. RL, on the other
hand, requires the learning objective to be encoded as scalar reward signals.
Since doing such translations manually is both tedious and error-prone, a
number of techniques have been proposed to translate high-level objectives
(expressed in logic or automata formalism) to scalar rewards for discrete-time
Markov decision processes (MDPs). Unfortunately, no automatic translation
exists for CTMDPs.
We consider CTMDP environments against the learning objectives expressed as
omega-regular languages. Omega-regular languages generalize regular languages
to infinite-horizon specifications and can express properties given in popular
linear-time logic LTL. To accommodate the dense-time nature of CTMDPs, we
consider two different semantics of omega-regular objectives: 1) satisfaction
semantics where the goal of the learner is to maximize the probability of
spending positive time in the good states, and 2) expectation semantics where
the goal of the learner is to optimize the long-run expected average time spent
in the ``good states" of the automaton. We present an approach enabling correct
translation to scalar reward signals that can be readily used by off-the-shelf
RL algorithms for CTMDPs. We demonstrate the effectiveness of the proposed
algorithms by evaluating it on some popular CTMDP benchmarks with omega-regular
objectives.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:45:38 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Falah",
"Amin",
""
],
[
"Guha",
"Shibashis",
""
],
[
"Trivedi",
"Ashutosh",
""
]
] | TITLE: Reinforcement Learning for Omega-Regular Specifications on
Continuous-Time MDP
ABSTRACT: Continuous-time Markov decision processes (CTMDPs) are canonical models to
express sequential decision-making under dense-time and stochastic
environments. When the stochastic evolution of the environment is only
available via sampling, model-free reinforcement learning (RL) is the
algorithm-of-choice to compute optimal decision sequence. RL, on the other
hand, requires the learning objective to be encoded as scalar reward signals.
Since doing such translations manually is both tedious and error-prone, a
number of techniques have been proposed to translate high-level objectives
(expressed in logic or automata formalism) to scalar rewards for discrete-time
Markov decision processes (MDPs). Unfortunately, no automatic translation
exists for CTMDPs.
We consider CTMDP environments against the learning objectives expressed as
omega-regular languages. Omega-regular languages generalize regular languages
to infinite-horizon specifications and can express properties given in popular
linear-time logic LTL. To accommodate the dense-time nature of CTMDPs, we
consider two different semantics of omega-regular objectives: 1) satisfaction
semantics where the goal of the learner is to maximize the probability of
spending positive time in the good states, and 2) expectation semantics where
the goal of the learner is to optimize the long-run expected average time spent
in the ``good states" of the automaton. We present an approach enabling correct
translation to scalar reward signals that can be readily used by off-the-shelf
RL algorithms for CTMDPs. We demonstrate the effectiveness of the proposed
algorithms by evaluating it on some popular CTMDP benchmarks with omega-regular
objectives.
| no_new_dataset | 0.947962 |
2303.09531 | Xinwei Zhang | Xinwei Zhang, Mingyi Hong and Jie Chen | GLASU: A Communication-Efficient Algorithm for Federated Learning with
Vertically Distributed Graph Data | null | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Vertical federated learning (VFL) is a distributed learning paradigm, where
computing clients collectively train a model based on the partial features of
the same set of samples they possess. Current research on VFL focuses on the
case when samples are independent, but it rarely addresses an emerging scenario
when samples are interrelated through a graph. For graph-structured data, graph
neural networks (GNNs) are competitive machine learning models, but a naive
implementation in the VFL setting causes a significant communication overhead.
Moreover, the analysis of the training is faced with a challenge caused by the
biased stochastic gradients. In this paper, we propose a model splitting method
that splits a backbone GNN across the clients and the server and a
communication-efficient algorithm, GLASU, to train such a model. GLASU adopts
lazy aggregation and stale updates to skip aggregation when evaluating the
model and skip feature exchanges during training, greatly reducing
communication. We offer a theoretical analysis and conduct extensive numerical
experiments on real-world datasets, showing that the proposed algorithm
effectively trains a GNN model, whose performance matches that of the backbone
GNN when trained in a centralized manner.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:47:55 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Zhang",
"Xinwei",
""
],
[
"Hong",
"Mingyi",
""
],
[
"Chen",
"Jie",
""
]
] | TITLE: GLASU: A Communication-Efficient Algorithm for Federated Learning with
Vertically Distributed Graph Data
ABSTRACT: Vertical federated learning (VFL) is a distributed learning paradigm, where
computing clients collectively train a model based on the partial features of
the same set of samples they possess. Current research on VFL focuses on the
case when samples are independent, but it rarely addresses an emerging scenario
when samples are interrelated through a graph. For graph-structured data, graph
neural networks (GNNs) are competitive machine learning models, but a naive
implementation in the VFL setting causes a significant communication overhead.
Moreover, the analysis of the training is faced with a challenge caused by the
biased stochastic gradients. In this paper, we propose a model splitting method
that splits a backbone GNN across the clients and the server and a
communication-efficient algorithm, GLASU, to train such a model. GLASU adopts
lazy aggregation and stale updates to skip aggregation when evaluating the
model and skip feature exchanges during training, greatly reducing
communication. We offer a theoretical analysis and conduct extensive numerical
experiments on real-world datasets, showing that the proposed algorithm
effectively trains a GNN model, whose performance matches that of the backbone
GNN when trained in a centralized manner.
| no_new_dataset | 0.944995 |
2303.09534 | Mai Nishimura | Mai Nishimura, Shohei Nobuhara, Ko Nishino | InCrowdFormer: On-Ground Pedestrian World Model From Egocentric Views | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We introduce an on-ground Pedestrian World Model, a computational model that
can predict how pedestrians move around an observer in the crowd on the ground
plane, but from just the egocentric-views of the observer. Our model,
InCrowdFormer, fully leverages the Transformer architecture by modeling
pedestrian interaction and egocentric to top-down view transformation with
attention, and autoregressively predicts on-ground positions of a variable
number of people with an encoder-decoder architecture. We encode the
uncertainties arising from unknown pedestrian heights with latent codes to
predict the posterior distributions of pedestrian positions. We validate the
effectiveness of InCrowdFormer on a novel prediction benchmark of real
movements. The results show that InCrowdFormer accurately predicts the future
coordination of pedestrians. To the best of our knowledge, InCrowdFormer is the
first-of-its-kind pedestrian world model which we believe will benefit a wide
range of egocentric-view applications including crowd navigation, tracking, and
synthesis.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:51:02 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Nishimura",
"Mai",
""
],
[
"Nobuhara",
"Shohei",
""
],
[
"Nishino",
"Ko",
""
]
] | TITLE: InCrowdFormer: On-Ground Pedestrian World Model From Egocentric Views
ABSTRACT: We introduce an on-ground Pedestrian World Model, a computational model that
can predict how pedestrians move around an observer in the crowd on the ground
plane, but from just the egocentric-views of the observer. Our model,
InCrowdFormer, fully leverages the Transformer architecture by modeling
pedestrian interaction and egocentric to top-down view transformation with
attention, and autoregressively predicts on-ground positions of a variable
number of people with an encoder-decoder architecture. We encode the
uncertainties arising from unknown pedestrian heights with latent codes to
predict the posterior distributions of pedestrian positions. We validate the
effectiveness of InCrowdFormer on a novel prediction benchmark of real
movements. The results show that InCrowdFormer accurately predicts the future
coordination of pedestrians. To the best of our knowledge, InCrowdFormer is the
first-of-its-kind pedestrian world model which we believe will benefit a wide
range of egocentric-view applications including crowd navigation, tracking, and
synthesis.
| no_new_dataset | 0.94474 |
2303.09536 | Wele Gedara Chaminda Bandara | Wele Gedara Chaminda Bandara, Vishal M. Patel | Deep Metric Learning for Unsupervised Remote Sensing Change Detection | Code available at https://github.com/wgcban/Metric-CD | null | null | null | cs.CV cs.LG eess.IV | http://creativecommons.org/licenses/by/4.0/ | Remote Sensing Change Detection (RS-CD) aims to detect relevant changes from
Multi-Temporal Remote Sensing Images (MT-RSIs), which aids in various RS
applications such as land cover, land use, human development analysis, and
disaster response. The performance of existing RS-CD methods is attributed to
training on large annotated datasets. Furthermore, most of these models are
less transferable in the sense that the trained model often performs very
poorly when there is a domain gap between training and test datasets. This
paper proposes an unsupervised CD method based on deep metric learning that can
deal with both of these issues. Given an MT-RSI, the proposed method generates
corresponding change probability map by iteratively optimizing an unsupervised
CD loss without training it on a large dataset. Our unsupervised CD method
consists of two interconnected deep networks, namely Deep-Change Probability
Generator (D-CPG) and Deep-Feature Extractor (D-FE). The D-CPG is designed to
predict change and no change probability maps for a given MT-RSI, while D-FE is
used to extract deep features of MT-RSI that will be further used in the
proposed unsupervised CD loss. We use transfer learning capability to
initialize the parameters of D-FE. We iteratively optimize the parameters of
D-CPG and D-FE for a given MT-RSI by minimizing the proposed unsupervised
``similarity-dissimilarity loss''. This loss is motivated by the principle of
metric learning where we simultaneously maximize the distance between change
pair-wise pixels while minimizing the distance between no-change pair-wise
pixels in bi-temporal image domain and their deep feature domain. The
experiments conducted on three CD datasets show that our unsupervised CD method
achieves significant improvements over the state-of-the-art supervised and
unsupervised CD methods. Code available at https://github.com/wgcban/Metric-CD
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:52:45 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Bandara",
"Wele Gedara Chaminda",
""
],
[
"Patel",
"Vishal M.",
""
]
] | TITLE: Deep Metric Learning for Unsupervised Remote Sensing Change Detection
ABSTRACT: Remote Sensing Change Detection (RS-CD) aims to detect relevant changes from
Multi-Temporal Remote Sensing Images (MT-RSIs), which aids in various RS
applications such as land cover, land use, human development analysis, and
disaster response. The performance of existing RS-CD methods is attributed to
training on large annotated datasets. Furthermore, most of these models are
less transferable in the sense that the trained model often performs very
poorly when there is a domain gap between training and test datasets. This
paper proposes an unsupervised CD method based on deep metric learning that can
deal with both of these issues. Given an MT-RSI, the proposed method generates
corresponding change probability map by iteratively optimizing an unsupervised
CD loss without training it on a large dataset. Our unsupervised CD method
consists of two interconnected deep networks, namely Deep-Change Probability
Generator (D-CPG) and Deep-Feature Extractor (D-FE). The D-CPG is designed to
predict change and no change probability maps for a given MT-RSI, while D-FE is
used to extract deep features of MT-RSI that will be further used in the
proposed unsupervised CD loss. We use transfer learning capability to
initialize the parameters of D-FE. We iteratively optimize the parameters of
D-CPG and D-FE for a given MT-RSI by minimizing the proposed unsupervised
``similarity-dissimilarity loss''. This loss is motivated by the principle of
metric learning where we simultaneously maximize the distance between change
pair-wise pixels while minimizing the distance between no-change pair-wise
pixels in bi-temporal image domain and their deep feature domain. The
experiments conducted on three CD datasets show that our unsupervised CD method
achieves significant improvements over the state-of-the-art supervised and
unsupervised CD methods. Code available at https://github.com/wgcban/Metric-CD
| no_new_dataset | 0.944485 |
2303.09555 | Chuang Gan | Tsun-Hsuan Wang, Pingchuan Ma, Andrew Everett Spielberg, Zhou Xian,
Hao Zhang, Joshua B. Tenenbaum, Daniela Rus, Chuang Gan | SoftZoo: A Soft Robot Co-design Benchmark For Locomotion In Diverse
Environments | ICLR 2023. Project page:
https://sites.google.com/view/softzoo-iclr-2023 | null | null | null | cs.RO cs.AI cs.CV cs.GR cs.LG | http://creativecommons.org/publicdomain/zero/1.0/ | While significant research progress has been made in robot learning for
control, unique challenges arise when simultaneously co-optimizing morphology.
Existing work has typically been tailored for particular environments or
representations. In order to more fully understand inherent design and
performance tradeoffs and accelerate the development of new breeds of soft
robots, a comprehensive virtual platform with well-established tasks,
environments, and evaluation metrics is needed. In this work, we introduce
SoftZoo, a soft robot co-design platform for locomotion in diverse
environments. SoftZoo supports an extensive, naturally-inspired material set,
including the ability to simulate environments such as flat ground, desert,
wetland, clay, ice, snow, shallow water, and ocean. Further, it provides a
variety of tasks relevant for soft robotics, including fast locomotion, agile
turning, and path following, as well as differentiable design representations
for morphology and control. Combined, these elements form a feature-rich
platform for analysis and development of soft robot co-design algorithms. We
benchmark prevalent representations and co-design algorithms, and shed light on
1) the interplay between environment, morphology, and behavior; 2) the
importance of design space representations; 3) the ambiguity in muscle
formation and controller synthesis; and 4) the value of differentiable physics.
We envision that SoftZoo will serve as a standard platform and template an
approach toward the development of novel representations and algorithms for
co-designing soft robots' behavioral and morphological intelligence.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:59:50 GMT"
}
] | 2023-03-17T00:00:00 | [
[
"Wang",
"Tsun-Hsuan",
""
],
[
"Ma",
"Pingchuan",
""
],
[
"Spielberg",
"Andrew Everett",
""
],
[
"Xian",
"Zhou",
""
],
[
"Zhang",
"Hao",
""
],
[
"Tenenbaum",
"Joshua B.",
""
],
[
"Rus",
"Daniela",
""
],
[
"Gan",
"Chuang",
""
]
] | TITLE: SoftZoo: A Soft Robot Co-design Benchmark For Locomotion In Diverse
Environments
ABSTRACT: While significant research progress has been made in robot learning for
control, unique challenges arise when simultaneously co-optimizing morphology.
Existing work has typically been tailored for particular environments or
representations. In order to more fully understand inherent design and
performance tradeoffs and accelerate the development of new breeds of soft
robots, a comprehensive virtual platform with well-established tasks,
environments, and evaluation metrics is needed. In this work, we introduce
SoftZoo, a soft robot co-design platform for locomotion in diverse
environments. SoftZoo supports an extensive, naturally-inspired material set,
including the ability to simulate environments such as flat ground, desert,
wetland, clay, ice, snow, shallow water, and ocean. Further, it provides a
variety of tasks relevant for soft robotics, including fast locomotion, agile
turning, and path following, as well as differentiable design representations
for morphology and control. Combined, these elements form a feature-rich
platform for analysis and development of soft robot co-design algorithms. We
benchmark prevalent representations and co-design algorithms, and shed light on
1) the interplay between environment, morphology, and behavior; 2) the
importance of design space representations; 3) the ambiguity in muscle
formation and controller synthesis; and 4) the value of differentiable physics.
We envision that SoftZoo will serve as a standard platform and template an
approach toward the development of novel representations and algorithms for
co-designing soft robots' behavioral and morphological intelligence.
| no_new_dataset | 0.940079 |
2009.09521 | Yashesh Dhebar | Yashesh Dhebar, Kalyanmoy Deb, Subramanya Nageshrao, Ling Zhu and
Dimitar Filev | Towards Interpretable-AI Policies Induction using Evolutionary Nonlinear
Decision Trees for Discrete Action Systems | main paper: 12 pages (pages 1-12), Supplementary Document: 5 pages
(from pages 13-17). Video link: https://youtu.be/DByYWTQ6X3E | IEEE Transactions on Cybernetics, 23 June 2023 | 10.1109/TCYB.2022.3180664 | 35737627 | cs.LG cs.NE cs.SY eess.SY stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Black-box AI induction methods such as deep reinforcement learning (DRL) are
increasingly being used to find optimal policies for a given control task.
Although policies represented using a black-box AI are capable of efficiently
executing the underlying control task and achieving optimal closed-loop
performance, the developed control rules are often complex and neither
interpretable nor explainable. In this paper, we use a recently proposed
nonlinear decision-tree (NLDT) approach to find a hierarchical set of control
rules in an attempt to maximize the open-loop performance for approximating and
explaining the pre-trained black-box DRL (oracle) agent using the labelled
state-action dataset. Recent advances in nonlinear optimization approaches
using evolutionary computation facilitates finding a hierarchical set of
nonlinear control rules as a function of state variables using a
computationally fast bilevel optimization procedure at each node of the
proposed NLDT. Additionally, we propose a re-optimization procedure for
enhancing closed-loop performance of an already derived NLDT. We evaluate our
proposed methodologies (open and closed-loop NLDTs) on different control
problems having multiple discrete actions. In all these problems our proposed
approach is able to find relatively simple and interpretable rules involving
one to four non-linear terms per rule, while simultaneously achieving on par
closed-loop performance when compared to a trained black-box DRL agent. A
post-processing approach for simplifying the NLDT is also suggested. The
obtained results are inspiring as they suggest the replacement of complicated
black-box DRL policies involving thousands of parameters (making them
non-interpretable) with relatively simple interpretable policies. Results are
encouraging and motivating to pursue further applications of proposed approach
in solving more complex control tasks.
| [
{
"version": "v1",
"created": "Sun, 20 Sep 2020 20:41:57 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Apr 2021 17:28:51 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Dhebar",
"Yashesh",
""
],
[
"Deb",
"Kalyanmoy",
""
],
[
"Nageshrao",
"Subramanya",
""
],
[
"Zhu",
"Ling",
""
],
[
"Filev",
"Dimitar",
""
]
] | TITLE: Towards Interpretable-AI Policies Induction using Evolutionary Nonlinear
Decision Trees for Discrete Action Systems
ABSTRACT: Black-box AI induction methods such as deep reinforcement learning (DRL) are
increasingly being used to find optimal policies for a given control task.
Although policies represented using a black-box AI are capable of efficiently
executing the underlying control task and achieving optimal closed-loop
performance, the developed control rules are often complex and neither
interpretable nor explainable. In this paper, we use a recently proposed
nonlinear decision-tree (NLDT) approach to find a hierarchical set of control
rules in an attempt to maximize the open-loop performance for approximating and
explaining the pre-trained black-box DRL (oracle) agent using the labelled
state-action dataset. Recent advances in nonlinear optimization approaches
using evolutionary computation facilitates finding a hierarchical set of
nonlinear control rules as a function of state variables using a
computationally fast bilevel optimization procedure at each node of the
proposed NLDT. Additionally, we propose a re-optimization procedure for
enhancing closed-loop performance of an already derived NLDT. We evaluate our
proposed methodologies (open and closed-loop NLDTs) on different control
problems having multiple discrete actions. In all these problems our proposed
approach is able to find relatively simple and interpretable rules involving
one to four non-linear terms per rule, while simultaneously achieving on par
closed-loop performance when compared to a trained black-box DRL agent. A
post-processing approach for simplifying the NLDT is also suggested. The
obtained results are inspiring as they suggest the replacement of complicated
black-box DRL policies involving thousands of parameters (making them
non-interpretable) with relatively simple interpretable policies. Results are
encouraging and motivating to pursue further applications of proposed approach
in solving more complex control tasks.
| no_new_dataset | 0.944434 |
2101.07415 | Xingyou Song | Xingyou Song, Krzysztof Choromanski, Jack Parker-Holder, Yunhao Tang,
Qiuyi Zhang, Daiyi Peng, Deepali Jain, Wenbo Gao, Aldo Pacchiano, Tamas
Sarlos, Yuxiang Yang | ES-ENAS: Efficient Evolutionary Optimization for Large Hybrid Search
Spaces | Previously published at ICLR 2020 NAS Workshop. See
https://github.com/google-research/google-research/tree/master/es_enas for
associated code | null | null | null | cs.LG cs.NE cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we approach the problem of optimizing blackbox functions over
large hybrid search spaces consisting of both combinatorial and continuous
parameters. We demonstrate that previous evolutionary algorithms which rely on
mutation-based approaches, while flexible over combinatorial spaces, suffer
from a curse of dimensionality in high dimensional continuous spaces both
theoretically and empirically, which thus limits their scope over hybrid search
spaces as well. In order to combat this curse, we propose ES-ENAS, a simple and
modular joint optimization procedure combining the class of sample-efficient
smoothed gradient techniques, commonly known as Evolutionary Strategies (ES),
with combinatorial optimizers in a highly scalable and intuitive way, inspired
by the one-shot or supernet paradigm introduced in Efficient Neural
Architecture Search (ENAS). By doing so, we achieve significantly more sample
efficiency, which we empirically demonstrate over synthetic benchmarks, and are
further able to apply ES-ENAS for architecture search over popular RL
benchmarks.
| [
{
"version": "v1",
"created": "Tue, 19 Jan 2021 02:19:05 GMT"
},
{
"version": "v2",
"created": "Thu, 3 Jun 2021 23:48:45 GMT"
},
{
"version": "v3",
"created": "Tue, 21 Sep 2021 23:50:08 GMT"
},
{
"version": "v4",
"created": "Mon, 6 Dec 2021 21:45:20 GMT"
},
{
"version": "v5",
"created": "Thu, 28 Apr 2022 01:41:21 GMT"
},
{
"version": "v6",
"created": "Wed, 15 Mar 2023 15:05:53 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Song",
"Xingyou",
""
],
[
"Choromanski",
"Krzysztof",
""
],
[
"Parker-Holder",
"Jack",
""
],
[
"Tang",
"Yunhao",
""
],
[
"Zhang",
"Qiuyi",
""
],
[
"Peng",
"Daiyi",
""
],
[
"Jain",
"Deepali",
""
],
[
"Gao",
"Wenbo",
""
],
[
"Pacchiano",
"Aldo",
""
],
[
"Sarlos",
"Tamas",
""
],
[
"Yang",
"Yuxiang",
""
]
] | TITLE: ES-ENAS: Efficient Evolutionary Optimization for Large Hybrid Search
Spaces
ABSTRACT: In this paper, we approach the problem of optimizing blackbox functions over
large hybrid search spaces consisting of both combinatorial and continuous
parameters. We demonstrate that previous evolutionary algorithms which rely on
mutation-based approaches, while flexible over combinatorial spaces, suffer
from a curse of dimensionality in high dimensional continuous spaces both
theoretically and empirically, which thus limits their scope over hybrid search
spaces as well. In order to combat this curse, we propose ES-ENAS, a simple and
modular joint optimization procedure combining the class of sample-efficient
smoothed gradient techniques, commonly known as Evolutionary Strategies (ES),
with combinatorial optimizers in a highly scalable and intuitive way, inspired
by the one-shot or supernet paradigm introduced in Efficient Neural
Architecture Search (ENAS). By doing so, we achieve significantly more sample
efficiency, which we empirically demonstrate over synthetic benchmarks, and are
further able to apply ES-ENAS for architecture search over popular RL
benchmarks.
| no_new_dataset | 0.952264 |
2102.06808 | Piotr Kozakowski | Piotr Kozakowski, Miko{\l}aj Pacek, Piotr Mi{\l}o\'s | Planning and Learning Using Adaptive Entropy Tree Search | null | null | 10.1109/IJCNN55064.2022.9892556 | null | cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent breakthroughs in Artificial Intelligence have shown that the
combination of tree-based planning with deep learning can lead to superior
performance. We present Adaptive Entropy Tree Search (ANTS) - a novel algorithm
combining planning and learning in the maximum entropy paradigm. Through a
comprehensive suite of experiments on the Atari benchmark we show that ANTS
significantly outperforms PUCT, the planning component of the state-of-the-art
AlphaZero system. ANTS builds upon recent work on maximum entropy planning
methods - which however, as we show, fail in combination with learning. ANTS
resolves this issue to reach state-of-the-art performance. We further find that
ANTS exhibits superior robustness to different hyperparameter choices, compared
to the previous algorithms. We believe that the high performance and robustness
of ANTS can bring tree search planning one step closer to wide practical
adoption.
| [
{
"version": "v1",
"created": "Fri, 12 Feb 2021 22:54:24 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Sep 2021 13:21:24 GMT"
},
{
"version": "v3",
"created": "Tue, 14 Mar 2023 22:29:46 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Kozakowski",
"Piotr",
""
],
[
"Pacek",
"Mikołaj",
""
],
[
"Miłoś",
"Piotr",
""
]
] | TITLE: Planning and Learning Using Adaptive Entropy Tree Search
ABSTRACT: Recent breakthroughs in Artificial Intelligence have shown that the
combination of tree-based planning with deep learning can lead to superior
performance. We present Adaptive Entropy Tree Search (ANTS) - a novel algorithm
combining planning and learning in the maximum entropy paradigm. Through a
comprehensive suite of experiments on the Atari benchmark we show that ANTS
significantly outperforms PUCT, the planning component of the state-of-the-art
AlphaZero system. ANTS builds upon recent work on maximum entropy planning
methods - which however, as we show, fail in combination with learning. ANTS
resolves this issue to reach state-of-the-art performance. We further find that
ANTS exhibits superior robustness to different hyperparameter choices, compared
to the previous algorithms. We believe that the high performance and robustness
of ANTS can bring tree search planning one step closer to wide practical
adoption.
| no_new_dataset | 0.942029 |
2103.06370 | Govardana Sachithanandam Ramachandran | Govardana Sachithanandam Ramachandran, Kazuma Hashimoto, Caiming Xiong | Causal-aware Safe Policy Improvement for Task-oriented dialogue | null | null | 10.18653/v1/2022.acl-long.8 | null | cs.CL cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The recent success of reinforcement learning's (RL) in solving complex tasks
is most often attributed to its capacity to explore and exploit an environment
where it has been trained. Sample efficiency is usually not an issue since
cheap simulators are available to sample data on-policy. On the other hand,
task oriented dialogues are usually learnt from offline data collected using
human demonstrations. Collecting diverse demonstrations and annotating them is
expensive. Unfortunately, use of RL methods trained on off-policy data are
prone to issues of bias and generalization, which are further exacerbated by
stochasticity in human response and non-markovian belief state of a dialogue
management system. To this end, we propose a batch RL framework for task
oriented dialogue policy learning: causal aware safe policy improvement
(CASPI). This method gives guarantees on dialogue policy's performance and also
learns to shape rewards according to intentions behind human responses, rather
than just mimicking demonstration data; this couple with batch-RL helps overall
with sample efficiency of the framework. We demonstrate the effectiveness of
this framework on a dialogue-context-to-text Generation and end-to-end dialogue
task of the Multiwoz2.0 dataset. The proposed method outperforms the current
state of the art on these metrics, in both case. In the end-to-end case, our
method trained only on 10\% of the data was able to out perform current state
in three out of four evaluation metrics.
| [
{
"version": "v1",
"created": "Wed, 10 Mar 2021 22:34:28 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Ramachandran",
"Govardana Sachithanandam",
""
],
[
"Hashimoto",
"Kazuma",
""
],
[
"Xiong",
"Caiming",
""
]
] | TITLE: Causal-aware Safe Policy Improvement for Task-oriented dialogue
ABSTRACT: The recent success of reinforcement learning's (RL) in solving complex tasks
is most often attributed to its capacity to explore and exploit an environment
where it has been trained. Sample efficiency is usually not an issue since
cheap simulators are available to sample data on-policy. On the other hand,
task oriented dialogues are usually learnt from offline data collected using
human demonstrations. Collecting diverse demonstrations and annotating them is
expensive. Unfortunately, use of RL methods trained on off-policy data are
prone to issues of bias and generalization, which are further exacerbated by
stochasticity in human response and non-markovian belief state of a dialogue
management system. To this end, we propose a batch RL framework for task
oriented dialogue policy learning: causal aware safe policy improvement
(CASPI). This method gives guarantees on dialogue policy's performance and also
learns to shape rewards according to intentions behind human responses, rather
than just mimicking demonstration data; this couple with batch-RL helps overall
with sample efficiency of the framework. We demonstrate the effectiveness of
this framework on a dialogue-context-to-text Generation and end-to-end dialogue
task of the Multiwoz2.0 dataset. The proposed method outperforms the current
state of the art on these metrics, in both case. In the end-to-end case, our
method trained only on 10\% of the data was able to out perform current state
in three out of four evaluation metrics.
| no_new_dataset | 0.942823 |
2106.07452 | Saad Hamid | Saad Hamid, Sebastian Schulze, Michael A. Osborne, Stephen J. Roberts | Marginalising over Stationary Kernels with Bayesian Quadrature | null | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Marginalising over families of Gaussian Process kernels produces flexible
model classes with well-calibrated uncertainty estimates. Existing approaches
require likelihood evaluations of many kernels, rendering them prohibitively
expensive for larger datasets. We propose a Bayesian Quadrature scheme to make
this marginalisation more efficient and thereby more practical. Through use of
the maximum mean discrepancies between distributions, we define a kernel over
kernels that captures invariances between Spectral Mixture (SM) Kernels. Kernel
samples are selected by generalising an information-theoretic acquisition
function for warped Bayesian Quadrature. We show that our framework achieves
more accurate predictions with better calibrated uncertainty than
state-of-the-art baselines, especially when given limited (wall-clock) time
budgets.
| [
{
"version": "v1",
"created": "Mon, 14 Jun 2021 14:23:34 GMT"
},
{
"version": "v2",
"created": "Sun, 27 Mar 2022 11:18:56 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Mar 2023 15:57:38 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Hamid",
"Saad",
""
],
[
"Schulze",
"Sebastian",
""
],
[
"Osborne",
"Michael A.",
""
],
[
"Roberts",
"Stephen J.",
""
]
] | TITLE: Marginalising over Stationary Kernels with Bayesian Quadrature
ABSTRACT: Marginalising over families of Gaussian Process kernels produces flexible
model classes with well-calibrated uncertainty estimates. Existing approaches
require likelihood evaluations of many kernels, rendering them prohibitively
expensive for larger datasets. We propose a Bayesian Quadrature scheme to make
this marginalisation more efficient and thereby more practical. Through use of
the maximum mean discrepancies between distributions, we define a kernel over
kernels that captures invariances between Spectral Mixture (SM) Kernels. Kernel
samples are selected by generalising an information-theoretic acquisition
function for warped Bayesian Quadrature. We show that our framework achieves
more accurate predictions with better calibrated uncertainty than
state-of-the-art baselines, especially when given limited (wall-clock) time
budgets.
| no_new_dataset | 0.94887 |
2106.10076 | Rui Song | Rui Song, Xingbing Chen, Zelong Liu, Haining An, Zhiqi Zhang,
Xiaoguang Wang, Hao Xu | Label prompt for multi-label text classification | null | null | null | 21 | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | One of the key problems in multi-label text classification is how to take
advantage of the correlation among labels. However, it is very challenging to
directly model the correlations among labels in a complex and unknown label
space. In this paper, we propose a Label Mask multi-label text classification
model (LM-MTC), which is inspired by the idea of cloze questions of language
model. LM-MTC is able to capture implicit relationships among labels through
the powerful ability of pre-train language models. On the basis, we assign a
different token to each potential label, and randomly mask the token with a
certain probability to build a label based Masked Language Model (MLM). We
train the MTC and MLM together, further improving the generalization ability of
the model. A large number of experiments on multiple datasets demonstrate the
effectiveness of our method.
| [
{
"version": "v1",
"created": "Fri, 18 Jun 2021 11:54:33 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 12:19:58 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Song",
"Rui",
""
],
[
"Chen",
"Xingbing",
""
],
[
"Liu",
"Zelong",
""
],
[
"An",
"Haining",
""
],
[
"Zhang",
"Zhiqi",
""
],
[
"Wang",
"Xiaoguang",
""
],
[
"Xu",
"Hao",
""
]
] | TITLE: Label prompt for multi-label text classification
ABSTRACT: One of the key problems in multi-label text classification is how to take
advantage of the correlation among labels. However, it is very challenging to
directly model the correlations among labels in a complex and unknown label
space. In this paper, we propose a Label Mask multi-label text classification
model (LM-MTC), which is inspired by the idea of cloze questions of language
model. LM-MTC is able to capture implicit relationships among labels through
the powerful ability of pre-train language models. On the basis, we assign a
different token to each potential label, and randomly mask the token with a
certain probability to build a label based Masked Language Model (MLM). We
train the MTC and MLM together, further improving the generalization ability of
the model. A large number of experiments on multiple datasets demonstrate the
effectiveness of our method.
| no_new_dataset | 0.949059 |
2202.13220 | Chen-Chou Lo | Chen-Chou Lo and Patrick Vandewalle | How Much Depth Information can Radar Contribute to a Depth Estimation
Model? | published on EI2023, 7 pages, 4 figures, 2 tables | null | 10.2352/EI.2023.35.16.AVM-122 | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | Recently, several works have proposed fusing radar data as an additional
perceptual signal into monocular depth estimation models because radar data is
robust against varying light and weather conditions. Although improved
performances were reported in prior works, it is still hard to tell how much
depth information radar can contribute to a depth estimation model. In this
paper, we propose radar inference and supervision experiments to investigate
the intrinsic depth potential of radar data using state-of-the-art depth
estimation models on the nuScenes dataset. In the inference experiment, the
model predicts depth by taking only radar as input to demonstrate the inference
capability using radar data. In the supervision experiment, a monocular depth
estimation model is trained under radar supervision to show the intrinsic depth
information that radar can contribute. Our experiments demonstrate that the
model using only sparse radar as input can detect the shape of surroundings to
a certain extent in the predicted depth. Furthermore, the monocular depth
estimation model supervised by preprocessed radar achieves a good performance
compared to the baseline model trained with sparse lidar supervision.
| [
{
"version": "v1",
"created": "Sat, 26 Feb 2022 20:02:47 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 10:26:32 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Lo",
"Chen-Chou",
""
],
[
"Vandewalle",
"Patrick",
""
]
] | TITLE: How Much Depth Information can Radar Contribute to a Depth Estimation
Model?
ABSTRACT: Recently, several works have proposed fusing radar data as an additional
perceptual signal into monocular depth estimation models because radar data is
robust against varying light and weather conditions. Although improved
performances were reported in prior works, it is still hard to tell how much
depth information radar can contribute to a depth estimation model. In this
paper, we propose radar inference and supervision experiments to investigate
the intrinsic depth potential of radar data using state-of-the-art depth
estimation models on the nuScenes dataset. In the inference experiment, the
model predicts depth by taking only radar as input to demonstrate the inference
capability using radar data. In the supervision experiment, a monocular depth
estimation model is trained under radar supervision to show the intrinsic depth
information that radar can contribute. Our experiments demonstrate that the
model using only sparse radar as input can detect the shape of surroundings to
a certain extent in the predicted depth. Furthermore, the monocular depth
estimation model supervised by preprocessed radar achieves a good performance
compared to the baseline model trained with sparse lidar supervision.
| no_new_dataset | 0.952264 |
2203.15901 | Alexandros Gkillas | Alexandros Gkillas, Dimitris Ampeliotis, Kostas Berberidis | Connections between Deep Equilibrium and Sparse Representation Models
with Application to Hyperspectral Image Denoising | null | IEEE Transactions on Image Processing ( Volume: 32) 22 February
2023 1513 - 1528 | 10.1109/TIP.2023.3245323 | null | cs.CV cs.NE eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this study, the problem of computing a sparse representation of
multi-dimensional visual data is considered. In general, such data e.g.,
hyperspectral images, color images or video data consists of signals that
exhibit strong local dependencies. A new computationally efficient sparse
coding optimization problem is derived by employing regularization terms that
are adapted to the properties of the signals of interest. Exploiting the merits
of the learnable regularization techniques, a neural network is employed to act
as structure prior and reveal the underlying signal dependencies. To solve the
optimization problem Deep unrolling and Deep equilibrium based algorithms are
developed, forming highly interpretable and concise deep-learning-based
architectures, that process the input dataset in a block-by-block fashion.
Extensive simulation results, in the context of hyperspectral image denoising,
are provided, which demonstrate that the proposed algorithms outperform
significantly other sparse coding approaches and exhibit superior performance
against recent state-of-the-art deep-learning-based denoising models. In a
wider perspective, our work provides a unique bridge between a classic
approach, that is the sparse representation theory, and modern representation
tools that are based on deep learning modeling.
| [
{
"version": "v1",
"created": "Tue, 29 Mar 2022 21:00:39 GMT"
},
{
"version": "v2",
"created": "Wed, 11 May 2022 20:28:27 GMT"
},
{
"version": "v3",
"created": "Tue, 14 Mar 2023 23:03:34 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Gkillas",
"Alexandros",
""
],
[
"Ampeliotis",
"Dimitris",
""
],
[
"Berberidis",
"Kostas",
""
]
] | TITLE: Connections between Deep Equilibrium and Sparse Representation Models
with Application to Hyperspectral Image Denoising
ABSTRACT: In this study, the problem of computing a sparse representation of
multi-dimensional visual data is considered. In general, such data e.g.,
hyperspectral images, color images or video data consists of signals that
exhibit strong local dependencies. A new computationally efficient sparse
coding optimization problem is derived by employing regularization terms that
are adapted to the properties of the signals of interest. Exploiting the merits
of the learnable regularization techniques, a neural network is employed to act
as structure prior and reveal the underlying signal dependencies. To solve the
optimization problem Deep unrolling and Deep equilibrium based algorithms are
developed, forming highly interpretable and concise deep-learning-based
architectures, that process the input dataset in a block-by-block fashion.
Extensive simulation results, in the context of hyperspectral image denoising,
are provided, which demonstrate that the proposed algorithms outperform
significantly other sparse coding approaches and exhibit superior performance
against recent state-of-the-art deep-learning-based denoising models. In a
wider perspective, our work provides a unique bridge between a classic
approach, that is the sparse representation theory, and modern representation
tools that are based on deep learning modeling.
| no_new_dataset | 0.942771 |
2204.04932 | Lizhou Liao | Lizhou Liao, Chunyun Fu, Binbin Feng, and Tian Su | Optimized SC-F-LOAM: Optimized Fast LiDAR Odometry and Mapping Using
Scan Context | null | Proceedings of the 2022 6th CAA International Conference on
Vehicular Control and Intelligence (CVCI), Nanjing, China, 28-30 October 2022 | 10.1109/CVCI56766.2022.9964574 | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | LiDAR odometry can achieve accurate vehicle pose estimation for short driving
range or in small-scale environments, but for long driving range or in
large-scale environments, the accuracy deteriorates as a result of cumulative
estimation errors. This drawback necessitates the inclusion of loop closure
detection in a SLAM framework to suppress the adverse effects of cumulative
errors. To improve the accuracy of pose estimation, we propose a new
LiDAR-based SLAM method which uses F-LOAM as LiDAR odometry, Scan Context for
loop closure detection, and GTSAM for global optimization. In our approach, an
adaptive distance threshold (instead of a fixed threshold) is employed for loop
closure detection, which achieves more accurate loop closure detection results.
Besides, a feature-based matching method is used in our approach to compute
vehicle pose transformations between loop closure point cloud pairs, instead of
using the raw point cloud obtained by the LiDAR sensor, which significantly
reduces the computation time. The KITTI dataset is used for verifications of
our method, and the experimental results demonstrate that the proposed method
outperforms typical LiDAR odometry/SLAM methods in the literature. Our code is
made publicly available for the benefit of the community.
| [
{
"version": "v1",
"created": "Mon, 11 Apr 2022 08:16:00 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 09:51:52 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Liao",
"Lizhou",
""
],
[
"Fu",
"Chunyun",
""
],
[
"Feng",
"Binbin",
""
],
[
"Su",
"Tian",
""
]
] | TITLE: Optimized SC-F-LOAM: Optimized Fast LiDAR Odometry and Mapping Using
Scan Context
ABSTRACT: LiDAR odometry can achieve accurate vehicle pose estimation for short driving
range or in small-scale environments, but for long driving range or in
large-scale environments, the accuracy deteriorates as a result of cumulative
estimation errors. This drawback necessitates the inclusion of loop closure
detection in a SLAM framework to suppress the adverse effects of cumulative
errors. To improve the accuracy of pose estimation, we propose a new
LiDAR-based SLAM method which uses F-LOAM as LiDAR odometry, Scan Context for
loop closure detection, and GTSAM for global optimization. In our approach, an
adaptive distance threshold (instead of a fixed threshold) is employed for loop
closure detection, which achieves more accurate loop closure detection results.
Besides, a feature-based matching method is used in our approach to compute
vehicle pose transformations between loop closure point cloud pairs, instead of
using the raw point cloud obtained by the LiDAR sensor, which significantly
reduces the computation time. The KITTI dataset is used for verifications of
our method, and the experimental results demonstrate that the proposed method
outperforms typical LiDAR odometry/SLAM methods in the literature. Our code is
made publicly available for the benefit of the community.
| no_new_dataset | 0.949902 |
2204.06242 | Arber Qoku | Arber Qoku and Florian Buettner | Encoding Domain Knowledge in Multi-view Latent Variable Models: A
Bayesian Approach with Structured Sparsity | 9 pages, 4 figures, 2 tables | null | null | null | stat.ML cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Many real-world systems are described not only by data from a single source
but via multiple data views. In genomic medicine, for instance, patients can be
characterized by data from different molecular layers. Latent variable models
with structured sparsity are a commonly used tool for disentangling variation
within and across data views. However, their interpretability is cumbersome
since it requires a direct inspection and interpretation of each factor from
domain experts. Here, we propose MuVI, a novel multi-view latent variable model
based on a modified horseshoe prior for modeling structured sparsity. This
facilitates the incorporation of limited and noisy domain knowledge, thereby
allowing for an analysis of multi-view data in an inherently explainable
manner. We demonstrate that our model (i) outperforms state-of-the-art
approaches for modeling structured sparsity in terms of the reconstruction
error and the precision/recall, (ii) robustly integrates noisy domain expertise
in the form of feature sets, (iii) promotes the identifiability of factors and
(iv) infers interpretable and biologically meaningful axes of variation in a
real-world multi-view dataset of cancer patients.
| [
{
"version": "v1",
"created": "Wed, 13 Apr 2022 08:22:31 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 14:06:40 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Qoku",
"Arber",
""
],
[
"Buettner",
"Florian",
""
]
] | TITLE: Encoding Domain Knowledge in Multi-view Latent Variable Models: A
Bayesian Approach with Structured Sparsity
ABSTRACT: Many real-world systems are described not only by data from a single source
but via multiple data views. In genomic medicine, for instance, patients can be
characterized by data from different molecular layers. Latent variable models
with structured sparsity are a commonly used tool for disentangling variation
within and across data views. However, their interpretability is cumbersome
since it requires a direct inspection and interpretation of each factor from
domain experts. Here, we propose MuVI, a novel multi-view latent variable model
based on a modified horseshoe prior for modeling structured sparsity. This
facilitates the incorporation of limited and noisy domain knowledge, thereby
allowing for an analysis of multi-view data in an inherently explainable
manner. We demonstrate that our model (i) outperforms state-of-the-art
approaches for modeling structured sparsity in terms of the reconstruction
error and the precision/recall, (ii) robustly integrates noisy domain expertise
in the form of feature sets, (iii) promotes the identifiability of factors and
(iv) infers interpretable and biologically meaningful axes of variation in a
real-world multi-view dataset of cancer patients.
| no_new_dataset | 0.942981 |
2205.12117 | Liang Xu | Liang Xu, Yi Cheng, Fan Zhang, Bingxuan Wu, Pengfei Shao, Peng Liu,
Shuwei Shen, Peng Yao, Ronald X.Xu | Phased Progressive Learning with Coupling-Regulation-Imbalance Loss for
Imbalanced Data Classification | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep convolutional neural networks often perform poorly when faced with
datasets that suffer from quantity imbalances and classification difficulties.
Despite advances in the field, existing two-stage approaches still exhibit
dataset bias or domain shift. To counter this, a phased progressive learning
schedule has been proposed that gradually shifts the emphasis from
representation learning to training the upper classifier. This approach is
particularly beneficial for datasets with larger imbalances or fewer samples.
Another new method a coupling-regulation-imbalance loss function is proposed,
which combines three parts: a correction term, Focal loss, and LDAM loss. This
loss is effective in addressing quantity imbalances and outliers, while
regulating the focus of attention on samples with varying classification
difficulties. These approaches have yielded satisfactory results on several
benchmark datasets, including Imbalanced CIFAR10, Imbalanced CIFAR100,
ImageNet-LT, and iNaturalist 2018, and can be easily generalized to other
imbalanced classification models.
| [
{
"version": "v1",
"created": "Tue, 24 May 2022 14:46:39 GMT"
},
{
"version": "v2",
"created": "Fri, 23 Sep 2022 09:04:05 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Mar 2023 13:10:32 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Xu",
"Liang",
""
],
[
"Cheng",
"Yi",
""
],
[
"Zhang",
"Fan",
""
],
[
"Wu",
"Bingxuan",
""
],
[
"Shao",
"Pengfei",
""
],
[
"Liu",
"Peng",
""
],
[
"Shen",
"Shuwei",
""
],
[
"Yao",
"Peng",
""
],
[
"Xu",
"Ronald X.",
""
]
] | TITLE: Phased Progressive Learning with Coupling-Regulation-Imbalance Loss for
Imbalanced Data Classification
ABSTRACT: Deep convolutional neural networks often perform poorly when faced with
datasets that suffer from quantity imbalances and classification difficulties.
Despite advances in the field, existing two-stage approaches still exhibit
dataset bias or domain shift. To counter this, a phased progressive learning
schedule has been proposed that gradually shifts the emphasis from
representation learning to training the upper classifier. This approach is
particularly beneficial for datasets with larger imbalances or fewer samples.
Another new method a coupling-regulation-imbalance loss function is proposed,
which combines three parts: a correction term, Focal loss, and LDAM loss. This
loss is effective in addressing quantity imbalances and outliers, while
regulating the focus of attention on samples with varying classification
difficulties. These approaches have yielded satisfactory results on several
benchmark datasets, including Imbalanced CIFAR10, Imbalanced CIFAR100,
ImageNet-LT, and iNaturalist 2018, and can be easily generalized to other
imbalanced classification models.
| no_new_dataset | 0.950227 |
2206.01310 | Aur\'elien Decelle | Nicolas B\'ereux, Aur\'elien Decelle, Cyril Furtlehner, Beatriz Seoane | Learning a Restricted Boltzmann Machine using biased Monte Carlo
sampling | 24 pages, 12 figures | SciPost Phys. 14, 032 (2023) | 10.21468/SciPostPhys.14.3.032 | null | cs.LG cond-mat.dis-nn cond-mat.stat-mech physics.comp-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Restricted Boltzmann Machines are simple and powerful generative models that
can encode any complex dataset. Despite all their advantages, in practice the
trainings are often unstable and it is difficult to assess their quality
because the dynamics are affected by extremely slow time dependencies. This
situation becomes critical when dealing with low-dimensional clustered
datasets, where the time required to sample ergodically the trained models
becomes computationally prohibitive. In this work, we show that this divergence
of Monte Carlo mixing times is related to a phenomenon of phase coexistence,
similar to that which occurs in physics near a first-order phase transition. We
show that sampling the equilibrium distribution using the Markov chain Monte
Carlo method can be dramatically accelerated when using biased sampling
techniques, in particular the Tethered Monte Carlo (TMC) method. This sampling
technique efficiently solves the problem of evaluating the quality of a given
trained model and generating new samples in a reasonable amount of time.
Moreover, we show that this sampling technique can also be used to improve the
computation of the log-likelihood gradient during training, leading to dramatic
improvements in training RBMs with artificial clustered datasets. On real
low-dimensional datasets, this new training method fits RBM models with
significantly faster relaxation dynamics than those obtained with standard PCD
recipes. We also show that TMC sampling can be used to recover the free-energy
profile of the RBM. This proves to be extremely useful to compute the
probability distribution of a given model and to improve the generation of new
decorrelated samples in slow PCD-trained models.
| [
{
"version": "v1",
"created": "Thu, 2 Jun 2022 21:29:01 GMT"
},
{
"version": "v2",
"created": "Wed, 5 Oct 2022 12:52:21 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Béreux",
"Nicolas",
""
],
[
"Decelle",
"Aurélien",
""
],
[
"Furtlehner",
"Cyril",
""
],
[
"Seoane",
"Beatriz",
""
]
] | TITLE: Learning a Restricted Boltzmann Machine using biased Monte Carlo
sampling
ABSTRACT: Restricted Boltzmann Machines are simple and powerful generative models that
can encode any complex dataset. Despite all their advantages, in practice the
trainings are often unstable and it is difficult to assess their quality
because the dynamics are affected by extremely slow time dependencies. This
situation becomes critical when dealing with low-dimensional clustered
datasets, where the time required to sample ergodically the trained models
becomes computationally prohibitive. In this work, we show that this divergence
of Monte Carlo mixing times is related to a phenomenon of phase coexistence,
similar to that which occurs in physics near a first-order phase transition. We
show that sampling the equilibrium distribution using the Markov chain Monte
Carlo method can be dramatically accelerated when using biased sampling
techniques, in particular the Tethered Monte Carlo (TMC) method. This sampling
technique efficiently solves the problem of evaluating the quality of a given
trained model and generating new samples in a reasonable amount of time.
Moreover, we show that this sampling technique can also be used to improve the
computation of the log-likelihood gradient during training, leading to dramatic
improvements in training RBMs with artificial clustered datasets. On real
low-dimensional datasets, this new training method fits RBM models with
significantly faster relaxation dynamics than those obtained with standard PCD
recipes. We also show that TMC sampling can be used to recover the free-energy
profile of the RBM. This proves to be extremely useful to compute the
probability distribution of a given model and to improve the generation of new
decorrelated samples in slow PCD-trained models.
| no_new_dataset | 0.949902 |
2206.11995 | Duc Nguyen | Duc Nguyen | Efficient and Accurate Top-$K$ Recovery from Choice Data | null | Proceedings of Machine Learning Research 180 PMLR 2022 page
1509:1518 | null | null | cs.LG cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The intersection of learning to rank and choice modeling is an active area of
research with applications in e-commerce, information retrieval and the social
sciences. In some applications such as recommendation systems, the statistician
is primarily interested in recovering the set of the top ranked items from a
large pool of items as efficiently as possible using passively collected
discrete choice data, i.e., the user picks one item from a set of multiple
items. Motivated by this practical consideration, we propose the choice-based
Borda count algorithm as a fast and accurate ranking algorithm for top
$K$-recovery i.e., correctly identifying all of the top $K$ items. We show that
the choice-based Borda count algorithm has optimal sample complexity for
top-$K$ recovery under a broad class of random utility models. We prove that in
the limit, the choice-based Borda count algorithm produces the same top-$K$
estimate as the commonly used Maximum Likelihood Estimate method but the
former's speed and simplicity brings considerable advantages in practice.
Experiments on both synthetic and real datasets show that the counting
algorithm is competitive with commonly used ranking algorithms in terms of
accuracy while being several orders of magnitude faster.
| [
{
"version": "v1",
"created": "Thu, 23 Jun 2022 22:05:08 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Nguyen",
"Duc",
""
]
] | TITLE: Efficient and Accurate Top-$K$ Recovery from Choice Data
ABSTRACT: The intersection of learning to rank and choice modeling is an active area of
research with applications in e-commerce, information retrieval and the social
sciences. In some applications such as recommendation systems, the statistician
is primarily interested in recovering the set of the top ranked items from a
large pool of items as efficiently as possible using passively collected
discrete choice data, i.e., the user picks one item from a set of multiple
items. Motivated by this practical consideration, we propose the choice-based
Borda count algorithm as a fast and accurate ranking algorithm for top
$K$-recovery i.e., correctly identifying all of the top $K$ items. We show that
the choice-based Borda count algorithm has optimal sample complexity for
top-$K$ recovery under a broad class of random utility models. We prove that in
the limit, the choice-based Borda count algorithm produces the same top-$K$
estimate as the commonly used Maximum Likelihood Estimate method but the
former's speed and simplicity brings considerable advantages in practice.
Experiments on both synthetic and real datasets show that the counting
algorithm is competitive with commonly used ranking algorithms in terms of
accuracy while being several orders of magnitude faster.
| no_new_dataset | 0.948106 |
2206.13386 | Olger Siebinga | Olger Siebinga, Arkady Zgonnikov, David Abbink | Uncovering variability in human driving behavior through automatic
extraction of similar traffic scenes from large naturalistic datasets | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Recently, multiple naturalistic traffic datasets of human-driven trajectories
have been published (e.g., highD, NGSim, and pNEUMA). These datasets have been
used in studies that investigate variability in human driving behavior, for
example for scenario-based validation of autonomous vehicle (AV) behavior,
modeling driver behavior, or validating driver models. Thus far, these studies
focused on the variability on an operational level (e.g., velocity profiles
during a lane change), not on a tactical level (i.e., to change lanes or not).
Investigating the variability on both levels is necessary to develop driver
models and AVs that include multiple tactical behaviors. To expose multi-level
variability, the human responses to the same traffic scene could be
investigated. However, no method exists to automatically extract similar scenes
from datasets. Here, we present a four-step extraction method that uses the
Hausdorff distance, a mathematical distance metric for sets. We performed a
case study on the highD dataset that showed that the method is practically
applicable. The human responses to the selected scenes exposed the variability
on both the tactical and operational levels. With this new method, the
variability in operational and tactical human behavior can be investigated,
without the need for costly and time-consuming driving-simulator experiments.
| [
{
"version": "v1",
"created": "Fri, 17 Jun 2022 06:59:51 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 07:43:31 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Siebinga",
"Olger",
""
],
[
"Zgonnikov",
"Arkady",
""
],
[
"Abbink",
"David",
""
]
] | TITLE: Uncovering variability in human driving behavior through automatic
extraction of similar traffic scenes from large naturalistic datasets
ABSTRACT: Recently, multiple naturalistic traffic datasets of human-driven trajectories
have been published (e.g., highD, NGSim, and pNEUMA). These datasets have been
used in studies that investigate variability in human driving behavior, for
example for scenario-based validation of autonomous vehicle (AV) behavior,
modeling driver behavior, or validating driver models. Thus far, these studies
focused on the variability on an operational level (e.g., velocity profiles
during a lane change), not on a tactical level (i.e., to change lanes or not).
Investigating the variability on both levels is necessary to develop driver
models and AVs that include multiple tactical behaviors. To expose multi-level
variability, the human responses to the same traffic scene could be
investigated. However, no method exists to automatically extract similar scenes
from datasets. Here, we present a four-step extraction method that uses the
Hausdorff distance, a mathematical distance metric for sets. We performed a
case study on the highD dataset that showed that the method is practically
applicable. The human responses to the selected scenes exposed the variability
on both the tactical and operational levels. With this new method, the
variability in operational and tactical human behavior can be investigated,
without the need for costly and time-consuming driving-simulator experiments.
| no_new_dataset | 0.934035 |
2207.02849 | Sang Keun Choe | Sang Keun Choe, Willie Neiswanger, Pengtao Xie, Eric Xing | Betty: An Automatic Differentiation Library for Multilevel Optimization | null | null | null | null | cs.LG cs.AI math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Gradient-based multilevel optimization (MLO) has gained attention as a
framework for studying numerous problems, ranging from hyperparameter
optimization and meta-learning to neural architecture search and reinforcement
learning. However, gradients in MLO, which are obtained by composing
best-response Jacobians via the chain rule, are notoriously difficult to
implement and memory/compute intensive. We take an initial step towards closing
this gap by introducing Betty, a software library for large-scale MLO. At its
core, we devise a novel dataflow graph for MLO, which allows us to (1) develop
efficient automatic differentiation for MLO that reduces the computational
complexity from O(d^3) to O(d^2), (2) incorporate systems support such as
mixed-precision and data-parallel training for scalability, and (3) facilitate
implementation of MLO programs of arbitrary complexity while allowing a modular
interface for diverse algorithmic and systems design choices. We empirically
demonstrate that Betty can be used to implement an array of MLO programs, while
also observing up to 11% increase in test accuracy, 14% decrease in GPU memory
usage, and 20% decrease in training wall time over existing implementations on
multiple benchmarks. We also showcase that Betty enables scaling MLO to models
with hundreds of millions of parameters. We open-source the code at
https://github.com/leopard-ai/betty.
| [
{
"version": "v1",
"created": "Tue, 5 Jul 2022 14:01:15 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 14:10:55 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Choe",
"Sang Keun",
""
],
[
"Neiswanger",
"Willie",
""
],
[
"Xie",
"Pengtao",
""
],
[
"Xing",
"Eric",
""
]
] | TITLE: Betty: An Automatic Differentiation Library for Multilevel Optimization
ABSTRACT: Gradient-based multilevel optimization (MLO) has gained attention as a
framework for studying numerous problems, ranging from hyperparameter
optimization and meta-learning to neural architecture search and reinforcement
learning. However, gradients in MLO, which are obtained by composing
best-response Jacobians via the chain rule, are notoriously difficult to
implement and memory/compute intensive. We take an initial step towards closing
this gap by introducing Betty, a software library for large-scale MLO. At its
core, we devise a novel dataflow graph for MLO, which allows us to (1) develop
efficient automatic differentiation for MLO that reduces the computational
complexity from O(d^3) to O(d^2), (2) incorporate systems support such as
mixed-precision and data-parallel training for scalability, and (3) facilitate
implementation of MLO programs of arbitrary complexity while allowing a modular
interface for diverse algorithmic and systems design choices. We empirically
demonstrate that Betty can be used to implement an array of MLO programs, while
also observing up to 11% increase in test accuracy, 14% decrease in GPU memory
usage, and 20% decrease in training wall time over existing implementations on
multiple benchmarks. We also showcase that Betty enables scaling MLO to models
with hundreds of millions of parameters. We open-source the code at
https://github.com/leopard-ai/betty.
| no_new_dataset | 0.942823 |
2208.12242 | Nataniel Ruiz | Nataniel Ruiz, Yuanzhen Li, Varun Jampani, Yael Pritch, Michael
Rubinstein and Kfir Aberman | DreamBooth: Fine Tuning Text-to-Image Diffusion Models for
Subject-Driven Generation | Published at CVPR 2023. Project page: https://dreambooth.github.io/ | null | null | null | cs.CV cs.GR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Large text-to-image models achieved a remarkable leap in the evolution of AI,
enabling high-quality and diverse synthesis of images from a given text prompt.
However, these models lack the ability to mimic the appearance of subjects in a
given reference set and synthesize novel renditions of them in different
contexts. In this work, we present a new approach for "personalization" of
text-to-image diffusion models. Given as input just a few images of a subject,
we fine-tune a pretrained text-to-image model such that it learns to bind a
unique identifier with that specific subject. Once the subject is embedded in
the output domain of the model, the unique identifier can be used to synthesize
novel photorealistic images of the subject contextualized in different scenes.
By leveraging the semantic prior embedded in the model with a new autogenous
class-specific prior preservation loss, our technique enables synthesizing the
subject in diverse scenes, poses, views and lighting conditions that do not
appear in the reference images. We apply our technique to several
previously-unassailable tasks, including subject recontextualization,
text-guided view synthesis, and artistic rendering, all while preserving the
subject's key features. We also provide a new dataset and evaluation protocol
for this new task of subject-driven generation. Project page:
https://dreambooth.github.io/
| [
{
"version": "v1",
"created": "Thu, 25 Aug 2022 17:45:49 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 17:52:27 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Ruiz",
"Nataniel",
""
],
[
"Li",
"Yuanzhen",
""
],
[
"Jampani",
"Varun",
""
],
[
"Pritch",
"Yael",
""
],
[
"Rubinstein",
"Michael",
""
],
[
"Aberman",
"Kfir",
""
]
] | TITLE: DreamBooth: Fine Tuning Text-to-Image Diffusion Models for
Subject-Driven Generation
ABSTRACT: Large text-to-image models achieved a remarkable leap in the evolution of AI,
enabling high-quality and diverse synthesis of images from a given text prompt.
However, these models lack the ability to mimic the appearance of subjects in a
given reference set and synthesize novel renditions of them in different
contexts. In this work, we present a new approach for "personalization" of
text-to-image diffusion models. Given as input just a few images of a subject,
we fine-tune a pretrained text-to-image model such that it learns to bind a
unique identifier with that specific subject. Once the subject is embedded in
the output domain of the model, the unique identifier can be used to synthesize
novel photorealistic images of the subject contextualized in different scenes.
By leveraging the semantic prior embedded in the model with a new autogenous
class-specific prior preservation loss, our technique enables synthesizing the
subject in diverse scenes, poses, views and lighting conditions that do not
appear in the reference images. We apply our technique to several
previously-unassailable tasks, including subject recontextualization,
text-guided view synthesis, and artistic rendering, all while preserving the
subject's key features. We also provide a new dataset and evaluation protocol
for this new task of subject-driven generation. Project page:
https://dreambooth.github.io/
| new_dataset | 0.935465 |
2209.03952 | Zhong-Qiu Wang | Zhong-Qiu Wang and Samuele Cornell and Shukjae Choi and Younglo Lee
and Byeong-Yeol Kim and Shinji Watanabe | TF-GridNet: Making Time-Frequency Domain Models Great Again for Monaural
Speaker Separation | in IEEE ICASSP 2023 | null | null | null | cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | We propose TF-GridNet, a novel multi-path deep neural network (DNN) operating
in the time-frequency (T-F) domain, for monaural talker-independent speaker
separation in anechoic conditions. The model stacks several multi-path blocks,
each consisting of an intra-frame spectral module, a sub-band temporal module,
and a full-band self-attention module, to leverage local and global
spectro-temporal information for separation. The model is trained to perform
complex spectral mapping, where the real and imaginary (RI) components of the
input mixture are stacked as input features to predict target RI components.
Besides using the scale-invariant signal-to-distortion ratio (SI-SDR) loss for
model training, we include a novel loss term to encourage separated sources to
add up to the input mixture. Without using dynamic mixing, we obtain 23.4 dB
SI-SDR improvement (SI-SDRi) on the WSJ0-2mix dataset, outperforming the
previous best by a large margin.
| [
{
"version": "v1",
"created": "Thu, 8 Sep 2022 17:56:35 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 16:01:40 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Wang",
"Zhong-Qiu",
""
],
[
"Cornell",
"Samuele",
""
],
[
"Choi",
"Shukjae",
""
],
[
"Lee",
"Younglo",
""
],
[
"Kim",
"Byeong-Yeol",
""
],
[
"Watanabe",
"Shinji",
""
]
] | TITLE: TF-GridNet: Making Time-Frequency Domain Models Great Again for Monaural
Speaker Separation
ABSTRACT: We propose TF-GridNet, a novel multi-path deep neural network (DNN) operating
in the time-frequency (T-F) domain, for monaural talker-independent speaker
separation in anechoic conditions. The model stacks several multi-path blocks,
each consisting of an intra-frame spectral module, a sub-band temporal module,
and a full-band self-attention module, to leverage local and global
spectro-temporal information for separation. The model is trained to perform
complex spectral mapping, where the real and imaginary (RI) components of the
input mixture are stacked as input features to predict target RI components.
Besides using the scale-invariant signal-to-distortion ratio (SI-SDR) loss for
model training, we include a novel loss term to encourage separated sources to
add up to the input mixture. Without using dynamic mixing, we obtain 23.4 dB
SI-SDR improvement (SI-SDRi) on the WSJ0-2mix dataset, outperforming the
previous best by a large margin.
| no_new_dataset | 0.943086 |
2209.10150 | Zhenhua Xu | Zhenhua Xu, Yuxuan Liu, Yuxiang Sun, Ming Liu, Lujia Wang | RNGDet++: Road Network Graph Detection by Transformer with Instance
Segmentation and Multi-scale Features Enhancement | Accepted by IEEE Robotics and Automation Letters (RA-L) | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The road network graph is a critical component for downstream tasks in
autonomous driving, such as global route planning and navigation. In the past
years, road network graphs are usually annotated by human experts manually,
which is time-consuming and labor-intensive. To annotate road network graphs
effectively and efficiently, automatic algorithms for road network graph
detection are demanded. Most existing methods either adopt a post-processing
step on semantic segmentation maps to produce road network graphs, or propose
graph-based algorithms to directly predict the graphs. However, these works
suffer from hard-coded algorithms and inferior performance. To enhance the
previous state-of-the-art (SOTA) method RNGDet, we add an instance segmentation
head to better supervise the training, and enable the network to leverage
multi-scale features of the backbone. Since the new proposed approach is
improved from RNGDet, we name it RNGDet++. Experimental results show that our
RNGDet++ outperforms baseline methods in terms of almost all evaluation metrics
on two large-scale public datasets. Our code and supplementary materials are
available at \url{https://tonyxuqaq.github.io/projects/RNGDetPlusPlus/}.
| [
{
"version": "v1",
"created": "Wed, 21 Sep 2022 07:06:46 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 03:05:12 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Xu",
"Zhenhua",
""
],
[
"Liu",
"Yuxuan",
""
],
[
"Sun",
"Yuxiang",
""
],
[
"Liu",
"Ming",
""
],
[
"Wang",
"Lujia",
""
]
] | TITLE: RNGDet++: Road Network Graph Detection by Transformer with Instance
Segmentation and Multi-scale Features Enhancement
ABSTRACT: The road network graph is a critical component for downstream tasks in
autonomous driving, such as global route planning and navigation. In the past
years, road network graphs are usually annotated by human experts manually,
which is time-consuming and labor-intensive. To annotate road network graphs
effectively and efficiently, automatic algorithms for road network graph
detection are demanded. Most existing methods either adopt a post-processing
step on semantic segmentation maps to produce road network graphs, or propose
graph-based algorithms to directly predict the graphs. However, these works
suffer from hard-coded algorithms and inferior performance. To enhance the
previous state-of-the-art (SOTA) method RNGDet, we add an instance segmentation
head to better supervise the training, and enable the network to leverage
multi-scale features of the backbone. Since the new proposed approach is
improved from RNGDet, we name it RNGDet++. Experimental results show that our
RNGDet++ outperforms baseline methods in terms of almost all evaluation metrics
on two large-scale public datasets. Our code and supplementary materials are
available at \url{https://tonyxuqaq.github.io/projects/RNGDetPlusPlus/}.
| no_new_dataset | 0.943452 |
2209.11934 | Bo Sun | Bo Sun, Lin Yang, Mohammad Hajiesmaili, Adam Wierman, John C.S. Lui,
Don Towsley, Danny H.K. Tsang | The Online Knapsack Problem with Departures | null | null | null | null | cs.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The online knapsack problem is a classic online resource allocation problem
in networking and operations research. Its basic version studies how to pack
online arriving items of different sizes and values into a capacity-limited
knapsack. In this paper, we study a general version that includes item
departures, while also considering multiple knapsacks and multi-dimensional
item sizes. We design a threshold-based online algorithm and prove that the
algorithm can achieve order-optimal competitive ratios. Beyond worst-case
performance guarantees, we also aim to achieve near-optimal average performance
under typical instances. Towards this goal, we propose a data-driven online
algorithm that learns within a policy-class that guarantees a worst-case
performance bound. In trace-driven experiments, we show that our data-driven
algorithm outperforms other benchmark algorithms in an application of online
knapsack to job scheduling for cloud computing.
| [
{
"version": "v1",
"created": "Sat, 24 Sep 2022 05:55:54 GMT"
},
{
"version": "v2",
"created": "Fri, 4 Nov 2022 15:18:26 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Mar 2023 06:22:51 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Sun",
"Bo",
""
],
[
"Yang",
"Lin",
""
],
[
"Hajiesmaili",
"Mohammad",
""
],
[
"Wierman",
"Adam",
""
],
[
"Lui",
"John C. S.",
""
],
[
"Towsley",
"Don",
""
],
[
"Tsang",
"Danny H. K.",
""
]
] | TITLE: The Online Knapsack Problem with Departures
ABSTRACT: The online knapsack problem is a classic online resource allocation problem
in networking and operations research. Its basic version studies how to pack
online arriving items of different sizes and values into a capacity-limited
knapsack. In this paper, we study a general version that includes item
departures, while also considering multiple knapsacks and multi-dimensional
item sizes. We design a threshold-based online algorithm and prove that the
algorithm can achieve order-optimal competitive ratios. Beyond worst-case
performance guarantees, we also aim to achieve near-optimal average performance
under typical instances. Towards this goal, we propose a data-driven online
algorithm that learns within a policy-class that guarantees a worst-case
performance bound. In trace-driven experiments, we show that our data-driven
algorithm outperforms other benchmark algorithms in an application of online
knapsack to job scheduling for cloud computing.
| no_new_dataset | 0.941061 |
2209.14261 | Peter Mitrano | Peter Mitrano, Alex LaGrassa, Oliver Kroemer, Dmitry Berenson | Focused Adaptation of Dynamics Models for Deformable Object Manipulation | Project Website:
https://sites.google.com/view/focused-adaptation-dynamics/home | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | In order to efficiently learn a dynamics model for a task in a new
environment, one can adapt a model learned in a similar source environment.
However, existing adaptation methods can fail when the target dataset contains
transitions where the dynamics are very different from the source environment.
For example, the source environment dynamics could be of a rope manipulated in
free-space, whereas the target dynamics could involve collisions and
deformation on obstacles. Our key insight is to improve data efficiency by
focusing model adaptation on only the regions where the source and target
dynamics are similar. In the rope example, adapting the free-space dynamics
requires significantly fewer data than adapting the free-space dynamics while
also learning collision dynamics. We propose a new method for adaptation that
is effective in adapting to regions of similar dynamics. Additionally, we
combine this adaptation method with prior work on planning with unreliable
dynamics to make a method for data-efficient online adaptation, called FOCUS.
We first demonstrate that the proposed adaptation method achieves statistically
significantly lower prediction error in regions of similar dynamics on
simulated rope manipulation and plant watering tasks. We then show on a
bimanual rope manipulation task that FOCUS achieves data-efficient online
learning, in simulation and in the real world.
| [
{
"version": "v1",
"created": "Wed, 28 Sep 2022 17:23:36 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 13:41:55 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Mitrano",
"Peter",
""
],
[
"LaGrassa",
"Alex",
""
],
[
"Kroemer",
"Oliver",
""
],
[
"Berenson",
"Dmitry",
""
]
] | TITLE: Focused Adaptation of Dynamics Models for Deformable Object Manipulation
ABSTRACT: In order to efficiently learn a dynamics model for a task in a new
environment, one can adapt a model learned in a similar source environment.
However, existing adaptation methods can fail when the target dataset contains
transitions where the dynamics are very different from the source environment.
For example, the source environment dynamics could be of a rope manipulated in
free-space, whereas the target dynamics could involve collisions and
deformation on obstacles. Our key insight is to improve data efficiency by
focusing model adaptation on only the regions where the source and target
dynamics are similar. In the rope example, adapting the free-space dynamics
requires significantly fewer data than adapting the free-space dynamics while
also learning collision dynamics. We propose a new method for adaptation that
is effective in adapting to regions of similar dynamics. Additionally, we
combine this adaptation method with prior work on planning with unreliable
dynamics to make a method for data-efficient online adaptation, called FOCUS.
We first demonstrate that the proposed adaptation method achieves statistically
significantly lower prediction error in regions of similar dynamics on
simulated rope manipulation and plant watering tasks. We then show on a
bimanual rope manipulation task that FOCUS achieves data-efficient online
learning, in simulation and in the real world.
| no_new_dataset | 0.95253 |
2210.02226 | Manuel Szewc | Jernej F. Kamenik, Manuel Szewc | Null Hypothesis Test for Anomaly Detection | 10 pages, 3 figures, 1 Table. Matches published version at Physics
Letters B. All code is available at
https://github.com/ManuelSzewc/Null_Hypothesis_Test_for_Anomaly_Detection.
Comments welcome! | Physics Letters B, 840, 2023, 137836 | 10.1016/j.physletb.2023.137836 | null | hep-ph cs.LG hep-ex | http://creativecommons.org/licenses/by/4.0/ | We extend the use of Classification Without Labels for anomaly detection with
a hypothesis test designed to exclude the background-only hypothesis. By
testing for statistical independence of the two discriminating dataset regions,
we are able to exclude the background-only hypothesis without relying on fixed
anomaly score cuts or extrapolations of background estimates between regions.
The method relies on the assumption of conditional independence of anomaly
score features and dataset regions, which can be ensured using existing
decorrelation techniques. As a benchmark example, we consider the LHC Olympics
dataset where we show that mutual information represents a suitable test for
statistical independence and our method exhibits excellent and robust
performance at different signal fractions even in presence of realistic feature
correlations.
| [
{
"version": "v1",
"created": "Wed, 5 Oct 2022 13:03:55 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Oct 2022 10:33:47 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Mar 2023 12:17:54 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Kamenik",
"Jernej F.",
""
],
[
"Szewc",
"Manuel",
""
]
] | TITLE: Null Hypothesis Test for Anomaly Detection
ABSTRACT: We extend the use of Classification Without Labels for anomaly detection with
a hypothesis test designed to exclude the background-only hypothesis. By
testing for statistical independence of the two discriminating dataset regions,
we are able to exclude the background-only hypothesis without relying on fixed
anomaly score cuts or extrapolations of background estimates between regions.
The method relies on the assumption of conditional independence of anomaly
score features and dataset regions, which can be ensured using existing
decorrelation techniques. As a benchmark example, we consider the LHC Olympics
dataset where we show that mutual information represents a suitable test for
statistical independence and our method exhibits excellent and robust
performance at different signal fractions even in presence of realistic feature
correlations.
| no_new_dataset | 0.938067 |
2210.05669 | Saeed Saadatnejad | Saeed Saadatnejad, Ali Rasekh, Mohammadreza Mofayezi, Yasamin
Medghalchi, Sara Rajabzadeh, Taylor Mordan, Alexandre Alahi | A generic diffusion-based approach for 3D human pose prediction in the
wild | Accepted to ICRA 2023 | null | null | null | cs.CV cs.HC cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Predicting 3D human poses in real-world scenarios, also known as human pose
forecasting, is inevitably subject to noisy inputs arising from inaccurate 3D
pose estimations and occlusions. To address these challenges, we propose a
diffusion-based approach that can predict given noisy observations. We frame
the prediction task as a denoising problem, where both observation and
prediction are considered as a single sequence containing missing elements
(whether in the observation or prediction horizon). All missing elements are
treated as noise and denoised with our conditional diffusion model. To better
handle long-term forecasting horizon, we present a temporal cascaded diffusion
model. We demonstrate the benefits of our approach on four publicly available
datasets (Human3.6M, HumanEva-I, AMASS, and 3DPW), outperforming the
state-of-the-art. Additionally, we show that our framework is generic enough to
improve any 3D pose prediction model as a pre-processing step to repair their
inputs and a post-processing step to refine their outputs. The code is
available online: \url{https://github.com/vita-epfl/DePOSit}.
| [
{
"version": "v1",
"created": "Tue, 11 Oct 2022 17:59:54 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 09:29:59 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Saadatnejad",
"Saeed",
""
],
[
"Rasekh",
"Ali",
""
],
[
"Mofayezi",
"Mohammadreza",
""
],
[
"Medghalchi",
"Yasamin",
""
],
[
"Rajabzadeh",
"Sara",
""
],
[
"Mordan",
"Taylor",
""
],
[
"Alahi",
"Alexandre",
""
]
] | TITLE: A generic diffusion-based approach for 3D human pose prediction in the
wild
ABSTRACT: Predicting 3D human poses in real-world scenarios, also known as human pose
forecasting, is inevitably subject to noisy inputs arising from inaccurate 3D
pose estimations and occlusions. To address these challenges, we propose a
diffusion-based approach that can predict given noisy observations. We frame
the prediction task as a denoising problem, where both observation and
prediction are considered as a single sequence containing missing elements
(whether in the observation or prediction horizon). All missing elements are
treated as noise and denoised with our conditional diffusion model. To better
handle long-term forecasting horizon, we present a temporal cascaded diffusion
model. We demonstrate the benefits of our approach on four publicly available
datasets (Human3.6M, HumanEva-I, AMASS, and 3DPW), outperforming the
state-of-the-art. Additionally, we show that our framework is generic enough to
improve any 3D pose prediction model as a pre-processing step to repair their
inputs and a post-processing step to refine their outputs. The code is
available online: \url{https://github.com/vita-epfl/DePOSit}.
| no_new_dataset | 0.944485 |
2210.06575 | Qiyu Dai | Qiyu Dai, Yan Zhu, Yiran Geng, Ciyu Ruan, Jiazhao Zhang, He Wang | GraspNeRF: Multiview-based 6-DoF Grasp Detection for Transparent and
Specular Objects Using Generalizable NeRF | IEEE International Conference on Robotics and Automation (ICRA), 2023 | null | null | null | cs.RO cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we tackle 6-DoF grasp detection for transparent and specular
objects, which is an important yet challenging problem in vision-based robotic
systems, due to the failure of depth cameras in sensing their geometry. We, for
the first time, propose a multiview RGB-based 6-DoF grasp detection network,
GraspNeRF, that leverages the generalizable neural radiance field (NeRF) to
achieve material-agnostic object grasping in clutter. Compared to the existing
NeRF-based 3-DoF grasp detection methods that rely on densely captured input
images and time-consuming per-scene optimization, our system can perform
zero-shot NeRF construction with sparse RGB inputs and reliably detect 6-DoF
grasps, both in real-time. The proposed framework jointly learns generalizable
NeRF and grasp detection in an end-to-end manner, optimizing the scene
representation construction for the grasping. For training data, we generate a
large-scale photorealistic domain-randomized synthetic dataset of grasping in
cluttered tabletop scenes that enables direct transfer to the real world. Our
extensive experiments in synthetic and real-world environments demonstrate that
our method significantly outperforms all the baselines in all the experiments
while remaining in real-time. Project page can be found at
https://pku-epic.github.io/GraspNeRF
| [
{
"version": "v1",
"created": "Wed, 12 Oct 2022 20:31:23 GMT"
},
{
"version": "v2",
"created": "Tue, 7 Mar 2023 07:26:40 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Mar 2023 17:35:57 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Dai",
"Qiyu",
""
],
[
"Zhu",
"Yan",
""
],
[
"Geng",
"Yiran",
""
],
[
"Ruan",
"Ciyu",
""
],
[
"Zhang",
"Jiazhao",
""
],
[
"Wang",
"He",
""
]
] | TITLE: GraspNeRF: Multiview-based 6-DoF Grasp Detection for Transparent and
Specular Objects Using Generalizable NeRF
ABSTRACT: In this work, we tackle 6-DoF grasp detection for transparent and specular
objects, which is an important yet challenging problem in vision-based robotic
systems, due to the failure of depth cameras in sensing their geometry. We, for
the first time, propose a multiview RGB-based 6-DoF grasp detection network,
GraspNeRF, that leverages the generalizable neural radiance field (NeRF) to
achieve material-agnostic object grasping in clutter. Compared to the existing
NeRF-based 3-DoF grasp detection methods that rely on densely captured input
images and time-consuming per-scene optimization, our system can perform
zero-shot NeRF construction with sparse RGB inputs and reliably detect 6-DoF
grasps, both in real-time. The proposed framework jointly learns generalizable
NeRF and grasp detection in an end-to-end manner, optimizing the scene
representation construction for the grasping. For training data, we generate a
large-scale photorealistic domain-randomized synthetic dataset of grasping in
cluttered tabletop scenes that enables direct transfer to the real world. Our
extensive experiments in synthetic and real-world environments demonstrate that
our method significantly outperforms all the baselines in all the experiments
while remaining in real-time. Project page can be found at
https://pku-epic.github.io/GraspNeRF
| new_dataset | 0.944331 |
2210.07179 | Oscar Ma\~nas | Oscar Ma\~nas, Pau Rodriguez, Saba Ahmadi, Aida Nematzadeh, Yash Goyal
and Aishwarya Agrawal | MAPL: Parameter-Efficient Adaptation of Unimodal Pre-Trained Models for
Vision-Language Few-Shot Prompting | Accepted at EACL 2023 (main track); 26 pages, 21 figures, 6 tables;
Pau Rodriguez and Saba Ahmadi had equal contributions | null | null | null | cs.CV cs.AI cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large pre-trained models have proved to be remarkable zero- and
(prompt-based) few-shot learners in unimodal vision and language tasks. We
propose MAPL, a simple and parameter-efficient method that reuses frozen
pre-trained unimodal models and leverages their strong generalization
capabilities in multimodal vision-language (VL) settings. MAPL learns a
lightweight mapping between the representation spaces of unimodal models using
aligned image-text data, and can generalize to unseen VL tasks from just a few
in-context examples. The small number of trainable parameters makes MAPL
effective at low-data and in-domain learning. Moreover, MAPL's modularity
enables easy extension to other pre-trained models. Extensive experiments on
several visual question answering and image captioning benchmarks show that
MAPL achieves superior or competitive performance compared to similar methods
while training orders of magnitude fewer parameters. MAPL can be trained in
just a few hours using modest computational resources and public datasets. We
release our code and pre-trained model weights at
https://github.com/mair-lab/mapl.
| [
{
"version": "v1",
"created": "Thu, 13 Oct 2022 17:02:23 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 00:02:06 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Mañas",
"Oscar",
""
],
[
"Rodriguez",
"Pau",
""
],
[
"Ahmadi",
"Saba",
""
],
[
"Nematzadeh",
"Aida",
""
],
[
"Goyal",
"Yash",
""
],
[
"Agrawal",
"Aishwarya",
""
]
] | TITLE: MAPL: Parameter-Efficient Adaptation of Unimodal Pre-Trained Models for
Vision-Language Few-Shot Prompting
ABSTRACT: Large pre-trained models have proved to be remarkable zero- and
(prompt-based) few-shot learners in unimodal vision and language tasks. We
propose MAPL, a simple and parameter-efficient method that reuses frozen
pre-trained unimodal models and leverages their strong generalization
capabilities in multimodal vision-language (VL) settings. MAPL learns a
lightweight mapping between the representation spaces of unimodal models using
aligned image-text data, and can generalize to unseen VL tasks from just a few
in-context examples. The small number of trainable parameters makes MAPL
effective at low-data and in-domain learning. Moreover, MAPL's modularity
enables easy extension to other pre-trained models. Extensive experiments on
several visual question answering and image captioning benchmarks show that
MAPL achieves superior or competitive performance compared to similar methods
while training orders of magnitude fewer parameters. MAPL can be trained in
just a few hours using modest computational resources and public datasets. We
release our code and pre-trained model weights at
https://github.com/mair-lab/mapl.
| no_new_dataset | 0.945901 |
2210.07241 | Yanjie Ze | Yanjie Ze, Nicklas Hansen, Yinbo Chen, Mohit Jain, Xiaolong Wang | Visual Reinforcement Learning with Self-Supervised 3D Representations | Accepted in RA-L 2023 and IROS 2023. Project page:
https://yanjieze.com/3d4rl/ | null | null | null | cs.LG cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A prominent approach to visual Reinforcement Learning (RL) is to learn an
internal state representation using self-supervised methods, which has the
potential benefit of improved sample-efficiency and generalization through
additional learning signal and inductive biases. However, while the real world
is inherently 3D, prior efforts have largely been focused on leveraging 2D
computer vision techniques as auxiliary self-supervision. In this work, we
present a unified framework for self-supervised learning of 3D representations
for motor control. Our proposed framework consists of two phases: a pretraining
phase where a deep voxel-based 3D autoencoder is pretrained on a large
object-centric dataset, and a finetuning phase where the representation is
jointly finetuned together with RL on in-domain data. We empirically show that
our method enjoys improved sample efficiency in simulated manipulation tasks
compared to 2D representation learning methods. Additionally, our learned
policies transfer zero-shot to a real robot setup with only approximate
geometric correspondence, and successfully solve motor control tasks that
involve grasping and lifting from a single, uncalibrated RGB camera. Code and
videos are available at https://yanjieze.com/3d4rl/ .
| [
{
"version": "v1",
"created": "Thu, 13 Oct 2022 17:59:55 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 08:21:03 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Ze",
"Yanjie",
""
],
[
"Hansen",
"Nicklas",
""
],
[
"Chen",
"Yinbo",
""
],
[
"Jain",
"Mohit",
""
],
[
"Wang",
"Xiaolong",
""
]
] | TITLE: Visual Reinforcement Learning with Self-Supervised 3D Representations
ABSTRACT: A prominent approach to visual Reinforcement Learning (RL) is to learn an
internal state representation using self-supervised methods, which has the
potential benefit of improved sample-efficiency and generalization through
additional learning signal and inductive biases. However, while the real world
is inherently 3D, prior efforts have largely been focused on leveraging 2D
computer vision techniques as auxiliary self-supervision. In this work, we
present a unified framework for self-supervised learning of 3D representations
for motor control. Our proposed framework consists of two phases: a pretraining
phase where a deep voxel-based 3D autoencoder is pretrained on a large
object-centric dataset, and a finetuning phase where the representation is
jointly finetuned together with RL on in-domain data. We empirically show that
our method enjoys improved sample efficiency in simulated manipulation tasks
compared to 2D representation learning methods. Additionally, our learned
policies transfer zero-shot to a real robot setup with only approximate
geometric correspondence, and successfully solve motor control tasks that
involve grasping and lifting from a single, uncalibrated RGB camera. Code and
videos are available at https://yanjieze.com/3d4rl/ .
| no_new_dataset | 0.943295 |
2210.13576 | Guangzhi Sun | Evonne P.C. Lee, Guangzhi Sun, Chao Zhang, Philip C. Woodland | Spectral Clustering-aware Learning of Embeddings for Speaker Diarisation | To appear in ICASSP 2023, 5 pages | null | null | null | cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In speaker diarisation, speaker embedding extraction models often suffer from
the mismatch between their training loss functions and the speaker clustering
method. In this paper, we propose the method of spectral clustering-aware
learning of embeddings (SCALE) to address the mismatch. Specifically, besides
an angular prototype cal (AP) loss, SCALE uses a novel affinity matrix loss
which directly minimises the error between the affinity matrix estimated from
speaker embeddings and the reference. SCALE also includes p-percentile
thresholding and Gaussian blur as two important hyper-parameters for spectral
clustering in training. Experiments on the AMI dataset showed that speaker
embeddings obtained with SCALE achieved over 50% relative speaker error rate
reductions using oracle segmentation, and over 30% relative diarisation error
rate reductions using automatic segmentation when compared to a strong baseline
with the AP-loss-based speaker embeddings.
| [
{
"version": "v1",
"created": "Mon, 24 Oct 2022 19:55:07 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Mar 2023 23:04:08 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Lee",
"Evonne P. C.",
""
],
[
"Sun",
"Guangzhi",
""
],
[
"Zhang",
"Chao",
""
],
[
"Woodland",
"Philip C.",
""
]
] | TITLE: Spectral Clustering-aware Learning of Embeddings for Speaker Diarisation
ABSTRACT: In speaker diarisation, speaker embedding extraction models often suffer from
the mismatch between their training loss functions and the speaker clustering
method. In this paper, we propose the method of spectral clustering-aware
learning of embeddings (SCALE) to address the mismatch. Specifically, besides
an angular prototype cal (AP) loss, SCALE uses a novel affinity matrix loss
which directly minimises the error between the affinity matrix estimated from
speaker embeddings and the reference. SCALE also includes p-percentile
thresholding and Gaussian blur as two important hyper-parameters for spectral
clustering in training. Experiments on the AMI dataset showed that speaker
embeddings obtained with SCALE achieved over 50% relative speaker error rate
reductions using oracle segmentation, and over 30% relative diarisation error
rate reductions using automatic segmentation when compared to a strong baseline
with the AP-loss-based speaker embeddings.
| no_new_dataset | 0.950457 |
2210.15447 | Takaaki Saeki | Takaaki Saeki, Heiga Zen, Zhehuai Chen, Nobuyuki Morioka, Gary Wang,
Yu Zhang, Ankur Bapna, Andrew Rosenberg, Bhuvana Ramabhadran | Virtuoso: Massive Multilingual Speech-Text Joint Semi-Supervised
Learning for Text-To-Speech | To appear in ICASSP 2023 | null | null | null | cs.SD cs.CL eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper proposes Virtuoso, a massively multilingual speech-text joint
semi-supervised learning framework for text-to-speech synthesis (TTS) models.
Existing multilingual TTS typically supports tens of languages, which are a
small fraction of the thousands of languages in the world. One difficulty to
scale multilingual TTS to hundreds of languages is collecting high-quality
speech-text paired data in low-resource languages. This study extends Maestro,
a speech-text joint pretraining framework for automatic speech recognition
(ASR), to speech generation tasks. To train a TTS model from various types of
speech and text data, different training schemes are designed to handle
supervised (paired TTS and ASR data) and unsupervised (untranscribed speech and
unspoken text) datasets. Experimental evaluation shows that 1) multilingual TTS
models trained on Virtuoso can achieve significantly better naturalness and
intelligibility than baseline ones in seen languages, and 2) they can
synthesize reasonably intelligible and naturally sounding speech for unseen
languages where no high-quality paired TTS data is available.
| [
{
"version": "v1",
"created": "Thu, 27 Oct 2022 14:09:48 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 10:52:03 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Saeki",
"Takaaki",
""
],
[
"Zen",
"Heiga",
""
],
[
"Chen",
"Zhehuai",
""
],
[
"Morioka",
"Nobuyuki",
""
],
[
"Wang",
"Gary",
""
],
[
"Zhang",
"Yu",
""
],
[
"Bapna",
"Ankur",
""
],
[
"Rosenberg",
"Andrew",
""
],
[
"Ramabhadran",
"Bhuvana",
""
]
] | TITLE: Virtuoso: Massive Multilingual Speech-Text Joint Semi-Supervised
Learning for Text-To-Speech
ABSTRACT: This paper proposes Virtuoso, a massively multilingual speech-text joint
semi-supervised learning framework for text-to-speech synthesis (TTS) models.
Existing multilingual TTS typically supports tens of languages, which are a
small fraction of the thousands of languages in the world. One difficulty to
scale multilingual TTS to hundreds of languages is collecting high-quality
speech-text paired data in low-resource languages. This study extends Maestro,
a speech-text joint pretraining framework for automatic speech recognition
(ASR), to speech generation tasks. To train a TTS model from various types of
speech and text data, different training schemes are designed to handle
supervised (paired TTS and ASR data) and unsupervised (untranscribed speech and
unspoken text) datasets. Experimental evaluation shows that 1) multilingual TTS
models trained on Virtuoso can achieve significantly better naturalness and
intelligibility than baseline ones in seen languages, and 2) they can
synthesize reasonably intelligible and naturally sounding speech for unseen
languages where no high-quality paired TTS data is available.
| no_new_dataset | 0.949669 |
2210.15551 | Chen Tang | Chen Tang, Hongbo Zhang, Tyler Loakman, Chenghua Lin, Frank Guerin | Terminology-aware Medical Dialogue Generation | Accepted by ICASSP 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Medical dialogue generation aims to generate responses according to a history
of dialogue turns between doctors and patients. Unlike open-domain dialogue
generation, this requires background knowledge specific to the medical domain.
Existing generative frameworks for medical dialogue generation fall short of
incorporating domain-specific knowledge, especially with regard to medical
terminology. In this paper, we propose a novel framework to improve medical
dialogue generation by considering features centered on domain-specific
terminology. We leverage an attention mechanism to incorporate terminologically
centred features, and fill in the semantic gap between medical background
knowledge and common utterances by enforcing language models to learn
terminology representations with an auxiliary terminology recognition task.
Experimental results demonstrate the effectiveness of our approach, in which
our proposed framework outperforms SOTA language models. Additionally, we
provide a new dataset with medical terminology annotations to support the
research on medical dialogue generation. Our dataset and code are available at
https://github.com/tangg555/meddialog.
| [
{
"version": "v1",
"created": "Thu, 27 Oct 2022 15:41:46 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 00:02:07 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Tang",
"Chen",
""
],
[
"Zhang",
"Hongbo",
""
],
[
"Loakman",
"Tyler",
""
],
[
"Lin",
"Chenghua",
""
],
[
"Guerin",
"Frank",
""
]
] | TITLE: Terminology-aware Medical Dialogue Generation
ABSTRACT: Medical dialogue generation aims to generate responses according to a history
of dialogue turns between doctors and patients. Unlike open-domain dialogue
generation, this requires background knowledge specific to the medical domain.
Existing generative frameworks for medical dialogue generation fall short of
incorporating domain-specific knowledge, especially with regard to medical
terminology. In this paper, we propose a novel framework to improve medical
dialogue generation by considering features centered on domain-specific
terminology. We leverage an attention mechanism to incorporate terminologically
centred features, and fill in the semantic gap between medical background
knowledge and common utterances by enforcing language models to learn
terminology representations with an auxiliary terminology recognition task.
Experimental results demonstrate the effectiveness of our approach, in which
our proposed framework outperforms SOTA language models. Additionally, we
provide a new dataset with medical terminology annotations to support the
research on medical dialogue generation. Our dataset and code are available at
https://github.com/tangg555/meddialog.
| new_dataset | 0.958924 |
2210.16554 | Guangzhi Sun | Guangzhi Sun, Chao Zhang, Philip C. Woodland | End-to-end Spoken Language Understanding with Tree-constrained Pointer
Generator | 5 pages, to appear in ICASSP 2023 | null | null | null | cs.CL cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | End-to-end spoken language understanding (SLU) suffers from the long-tail
word problem. This paper exploits contextual biasing, a technique to improve
the speech recognition of rare words, in end-to-end SLU systems. Specifically,
a tree-constrained pointer generator (TCPGen), a powerful and efficient biasing
model component, is studied, which leverages a slot shortlist with
corresponding entities to extract biasing lists. Meanwhile, to bias the SLU
model output slot distribution, a slot probability biasing (SPB) mechanism is
proposed to calculate a slot distribution from TCPGen. Experiments on the SLURP
dataset showed consistent SLU-F1 improvements using TCPGen and SPB, especially
on unseen entities. On a new split by holding out 5 slot types for the test,
TCPGen with SPB achieved zero-shot learning with an SLU-F1 score over 50%
compared to baselines which can not deal with it. In addition to slot filling,
the intent classification accuracy was also improved.
| [
{
"version": "v1",
"created": "Sat, 29 Oct 2022 10:03:56 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Mar 2023 22:57:12 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Sun",
"Guangzhi",
""
],
[
"Zhang",
"Chao",
""
],
[
"Woodland",
"Philip C.",
""
]
] | TITLE: End-to-end Spoken Language Understanding with Tree-constrained Pointer
Generator
ABSTRACT: End-to-end spoken language understanding (SLU) suffers from the long-tail
word problem. This paper exploits contextual biasing, a technique to improve
the speech recognition of rare words, in end-to-end SLU systems. Specifically,
a tree-constrained pointer generator (TCPGen), a powerful and efficient biasing
model component, is studied, which leverages a slot shortlist with
corresponding entities to extract biasing lists. Meanwhile, to bias the SLU
model output slot distribution, a slot probability biasing (SPB) mechanism is
proposed to calculate a slot distribution from TCPGen. Experiments on the SLURP
dataset showed consistent SLU-F1 improvements using TCPGen and SPB, especially
on unseen entities. On a new split by holding out 5 slot types for the test,
TCPGen with SPB achieved zero-shot learning with an SLU-F1 score over 50%
compared to baselines which can not deal with it. In addition to slot filling,
the intent classification accuracy was also improved.
| no_new_dataset | 0.946151 |
2211.05231 | Joachim Ott | Joachim Ott and Shih-Chii Liu | Biologically-Inspired Continual Learning of Human Motion Sequences | 5 pages, 2 figures, accepted at IEEE International Conference on
Acoustics, Speech and Signal Processing (ICASSP), Rhodes Island, Greece, 2023 | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work proposes a model for continual learning on tasks involving temporal
sequences, specifically, human motions. It improves on a recently proposed
brain-inspired replay model (BI-R) by building a biologically-inspired
conditional temporal variational autoencoder (BI-CTVAE), which instantiates a
latent mixture-of-Gaussians for class representation. We investigate a novel
continual-learning-to-generate (CL2Gen) scenario where the model generates
motion sequences of different classes. The generative accuracy of the model is
tested over a set of tasks. The final classification accuracy of BI-CTVAE on a
human motion dataset after sequentially learning all action classes is 78%,
which is 63% higher than using no-replay, and only 5.4% lower than a
state-of-the-art offline trained GRU model.
| [
{
"version": "v1",
"created": "Wed, 2 Nov 2022 17:58:47 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 14:31:46 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Ott",
"Joachim",
""
],
[
"Liu",
"Shih-Chii",
""
]
] | TITLE: Biologically-Inspired Continual Learning of Human Motion Sequences
ABSTRACT: This work proposes a model for continual learning on tasks involving temporal
sequences, specifically, human motions. It improves on a recently proposed
brain-inspired replay model (BI-R) by building a biologically-inspired
conditional temporal variational autoencoder (BI-CTVAE), which instantiates a
latent mixture-of-Gaussians for class representation. We investigate a novel
continual-learning-to-generate (CL2Gen) scenario where the model generates
motion sequences of different classes. The generative accuracy of the model is
tested over a set of tasks. The final classification accuracy of BI-CTVAE on a
human motion dataset after sequentially learning all action classes is 78%,
which is 63% higher than using no-replay, and only 5.4% lower than a
state-of-the-art offline trained GRU model.
| no_new_dataset | 0.949153 |
2211.06444 | Bashirul Azam Biswas | Bashirul Azam Biswas and Qiang Ji | Probabilistic Debiasing of Scene Graphs | Accepted at CVPR 2023. Code available at
https://github.com/bashirulazam/within-triplet-debias | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The quality of scene graphs generated by the state-of-the-art (SOTA) models
is compromised due to the long-tail nature of the relationships and their
parent object pairs. Training of the scene graphs is dominated by the majority
relationships of the majority pairs and, therefore, the object-conditional
distributions of relationship in the minority pairs are not preserved after the
training is converged. Consequently, the biased model performs well on more
frequent relationships in the marginal distribution of relationships such as
`on' and `wearing', and performs poorly on the less frequent relationships such
as `eating' or `hanging from'. In this work, we propose virtual evidence
incorporated within-triplet Bayesian Network (BN) to preserve the
object-conditional distribution of the relationship label and to eradicate the
bias created by the marginal probability of the relationships. The insufficient
number of relationships in the minority classes poses a significant problem in
learning the within-triplet Bayesian network. We address this insufficiency by
embedding-based augmentation of triplets where we borrow samples of the
minority triplet classes from its neighborhood triplets in the semantic space.
We perform experiments on two different datasets and achieve a significant
improvement in the mean recall of the relationships. We also achieve better
balance between recall and mean recall performance compared to the SOTA
de-biasing techniques of scene graph models.
| [
{
"version": "v1",
"created": "Fri, 11 Nov 2022 19:06:49 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Mar 2023 21:07:45 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Biswas",
"Bashirul Azam",
""
],
[
"Ji",
"Qiang",
""
]
] | TITLE: Probabilistic Debiasing of Scene Graphs
ABSTRACT: The quality of scene graphs generated by the state-of-the-art (SOTA) models
is compromised due to the long-tail nature of the relationships and their
parent object pairs. Training of the scene graphs is dominated by the majority
relationships of the majority pairs and, therefore, the object-conditional
distributions of relationship in the minority pairs are not preserved after the
training is converged. Consequently, the biased model performs well on more
frequent relationships in the marginal distribution of relationships such as
`on' and `wearing', and performs poorly on the less frequent relationships such
as `eating' or `hanging from'. In this work, we propose virtual evidence
incorporated within-triplet Bayesian Network (BN) to preserve the
object-conditional distribution of the relationship label and to eradicate the
bias created by the marginal probability of the relationships. The insufficient
number of relationships in the minority classes poses a significant problem in
learning the within-triplet Bayesian network. We address this insufficiency by
embedding-based augmentation of triplets where we borrow samples of the
minority triplet classes from its neighborhood triplets in the semantic space.
We perform experiments on two different datasets and achieve a significant
improvement in the mean recall of the relationships. We also achieve better
balance between recall and mean recall performance compared to the SOTA
de-biasing techniques of scene graph models.
| no_new_dataset | 0.952309 |
2211.10772 | Maoyuan Ye | Maoyuan Ye, Jing Zhang, Shanshan Zhao, Juhua Liu, Tongliang Liu, Bo
Du, Dacheng Tao | DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text
Spotting | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | End-to-end text spotting aims to integrate scene text detection and
recognition into a unified framework. Dealing with the relationship between the
two sub-tasks plays a pivotal role in designing effective spotters. Although
Transformer-based methods eliminate the heuristic post-processing, they still
suffer from the synergy issue between the sub-tasks and low training
efficiency. In this paper, we present DeepSolo, a simple DETR-like baseline
that lets a single Decoder with Explicit Points Solo for text detection and
recognition simultaneously. Technically, for each text instance, we represent
the character sequence as ordered points and model them with learnable explicit
point queries. After passing a single decoder, the point queries have encoded
requisite text semantics and locations, thus can be further decoded to the
center line, boundary, script, and confidence of text via very simple
prediction heads in parallel. Besides, we also introduce a text-matching
criterion to deliver more accurate supervisory signals, thus enabling more
efficient training. Quantitative experiments on public benchmarks demonstrate
that DeepSolo outperforms previous state-of-the-art methods and achieves better
training efficiency. In addition, DeepSolo is also compatible with line
annotations, which require much less annotation cost than polygons. The code is
available at https://github.com/ViTAE-Transformer/DeepSolo.
| [
{
"version": "v1",
"created": "Sat, 19 Nov 2022 19:06:22 GMT"
},
{
"version": "v2",
"created": "Wed, 23 Nov 2022 07:36:17 GMT"
},
{
"version": "v3",
"created": "Mon, 6 Mar 2023 13:36:16 GMT"
},
{
"version": "v4",
"created": "Wed, 15 Mar 2023 14:03:07 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Ye",
"Maoyuan",
""
],
[
"Zhang",
"Jing",
""
],
[
"Zhao",
"Shanshan",
""
],
[
"Liu",
"Juhua",
""
],
[
"Liu",
"Tongliang",
""
],
[
"Du",
"Bo",
""
],
[
"Tao",
"Dacheng",
""
]
] | TITLE: DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text
Spotting
ABSTRACT: End-to-end text spotting aims to integrate scene text detection and
recognition into a unified framework. Dealing with the relationship between the
two sub-tasks plays a pivotal role in designing effective spotters. Although
Transformer-based methods eliminate the heuristic post-processing, they still
suffer from the synergy issue between the sub-tasks and low training
efficiency. In this paper, we present DeepSolo, a simple DETR-like baseline
that lets a single Decoder with Explicit Points Solo for text detection and
recognition simultaneously. Technically, for each text instance, we represent
the character sequence as ordered points and model them with learnable explicit
point queries. After passing a single decoder, the point queries have encoded
requisite text semantics and locations, thus can be further decoded to the
center line, boundary, script, and confidence of text via very simple
prediction heads in parallel. Besides, we also introduce a text-matching
criterion to deliver more accurate supervisory signals, thus enabling more
efficient training. Quantitative experiments on public benchmarks demonstrate
that DeepSolo outperforms previous state-of-the-art methods and achieves better
training efficiency. In addition, DeepSolo is also compatible with line
annotations, which require much less annotation cost than polygons. The code is
available at https://github.com/ViTAE-Transformer/DeepSolo.
| no_new_dataset | 0.943348 |
2211.14865 | Sai Krishna Rallabandi | Parag Pravin Dakle, SaiKrishna Rallabandi and Preethi Raghavan | Understanding BLOOM: An empirical study on diverse NLP tasks | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | We view the landscape of large language models (LLMs) through the lens of the
recently released BLOOM model to understand the performance of BLOOM and other
decoder-only LLMs compared to BERT-style encoder-only models. We achieve this
by evaluating the smaller BLOOM model variants (\textit{350m/560m} and
\textit{1b3/1b7}) on several NLP benchmark datasets and popular leaderboards.
We make the following observations: (1) BLOOM performance does not scale with
parameter size, unlike other LLMs like GPT and BERT. Experiments fine-tuning
BLOOM models show that the 560m variant performs similarly to or better than
the 1b7 variant, (2) Zero-shot cross-lingual and multi-lingual fine-tuning
experiments show that BLOOM is at par or worse than monolingual GPT-2 models,
and (3) Toxicity analysis of prompt-based text generation using the
RealToxicityPrompts dataset shows that the text generated by BLOOM is at least
17\% less toxic than GPT-2 and GPT-3 models.
| [
{
"version": "v1",
"created": "Sun, 27 Nov 2022 15:48:14 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 03:54:14 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Dakle",
"Parag Pravin",
""
],
[
"Rallabandi",
"SaiKrishna",
""
],
[
"Raghavan",
"Preethi",
""
]
] | TITLE: Understanding BLOOM: An empirical study on diverse NLP tasks
ABSTRACT: We view the landscape of large language models (LLMs) through the lens of the
recently released BLOOM model to understand the performance of BLOOM and other
decoder-only LLMs compared to BERT-style encoder-only models. We achieve this
by evaluating the smaller BLOOM model variants (\textit{350m/560m} and
\textit{1b3/1b7}) on several NLP benchmark datasets and popular leaderboards.
We make the following observations: (1) BLOOM performance does not scale with
parameter size, unlike other LLMs like GPT and BERT. Experiments fine-tuning
BLOOM models show that the 560m variant performs similarly to or better than
the 1b7 variant, (2) Zero-shot cross-lingual and multi-lingual fine-tuning
experiments show that BLOOM is at par or worse than monolingual GPT-2 models,
and (3) Toxicity analysis of prompt-based text generation using the
RealToxicityPrompts dataset shows that the text generated by BLOOM is at least
17\% less toxic than GPT-2 and GPT-3 models.
| no_new_dataset | 0.941815 |
2211.15115 | An Wenbin | Wenbin An, Feng Tian, Qinghua Zheng, Wei Ding, QianYing Wang, Ping
Chen | Generalized Category Discovery with Decoupled Prototypical Network | Accepted by AAAI 2023 | null | null | null | cs.CL cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generalized Category Discovery (GCD) aims to recognize both known and novel
categories from a set of unlabeled data, based on another dataset labeled with
only known categories. Without considering differences between known and novel
categories, current methods learn about them in a coupled manner, which can
hurt model's generalization and discriminative ability. Furthermore, the
coupled training approach prevents these models transferring category-specific
knowledge explicitly from labeled data to unlabeled data, which can lose
high-level semantic information and impair model performance. To mitigate above
limitations, we present a novel model called Decoupled Prototypical Network
(DPN). By formulating a bipartite matching problem for category prototypes, DPN
can not only decouple known and novel categories to achieve different training
targets effectively, but also align known categories in labeled and unlabeled
data to transfer category-specific knowledge explicitly and capture high-level
semantics. Furthermore, DPN can learn more discriminative features for both
known and novel categories through our proposed Semantic-aware Prototypical
Learning (SPL). Besides capturing meaningful semantic information, SPL can also
alleviate the noise of hard pseudo labels through semantic-weighted soft
assignment. Extensive experiments show that DPN outperforms state-of-the-art
models by a large margin on all evaluation metrics across multiple benchmark
datasets. Code and data are available at https://github.com/Lackel/DPN.
| [
{
"version": "v1",
"created": "Mon, 28 Nov 2022 08:05:45 GMT"
},
{
"version": "v2",
"created": "Wed, 15 Mar 2023 07:35:09 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"An",
"Wenbin",
""
],
[
"Tian",
"Feng",
""
],
[
"Zheng",
"Qinghua",
""
],
[
"Ding",
"Wei",
""
],
[
"Wang",
"QianYing",
""
],
[
"Chen",
"Ping",
""
]
] | TITLE: Generalized Category Discovery with Decoupled Prototypical Network
ABSTRACT: Generalized Category Discovery (GCD) aims to recognize both known and novel
categories from a set of unlabeled data, based on another dataset labeled with
only known categories. Without considering differences between known and novel
categories, current methods learn about them in a coupled manner, which can
hurt model's generalization and discriminative ability. Furthermore, the
coupled training approach prevents these models transferring category-specific
knowledge explicitly from labeled data to unlabeled data, which can lose
high-level semantic information and impair model performance. To mitigate above
limitations, we present a novel model called Decoupled Prototypical Network
(DPN). By formulating a bipartite matching problem for category prototypes, DPN
can not only decouple known and novel categories to achieve different training
targets effectively, but also align known categories in labeled and unlabeled
data to transfer category-specific knowledge explicitly and capture high-level
semantics. Furthermore, DPN can learn more discriminative features for both
known and novel categories through our proposed Semantic-aware Prototypical
Learning (SPL). Besides capturing meaningful semantic information, SPL can also
alleviate the noise of hard pseudo labels through semantic-weighted soft
assignment. Extensive experiments show that DPN outperforms state-of-the-art
models by a large margin on all evaluation metrics across multiple benchmark
datasets. Code and data are available at https://github.com/Lackel/DPN.
| no_new_dataset | 0.94801 |
2211.15158 | Sein Kim | Sein Kim, Namkyeong Lee, Junseok Lee, Dongmin Hyun and Chanyoung Park | Heterogeneous Graph Learning for Multi-modal Medical Data Analysis | AAAI 2023 | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Routine clinical visits of a patient produce not only image data, but also
non-image data containing clinical information regarding the patient, i.e.,
medical data is multi-modal in nature. Such heterogeneous modalities offer
different and complementary perspectives on the same patient, resulting in more
accurate clinical decisions when they are properly combined. However, despite
its significance, how to effectively fuse the multi-modal medical data into a
unified framework has received relatively little attention. In this paper, we
propose an effective graph-based framework called HetMed (Heterogeneous Graph
Learning for Multi-modal Medical Data Analysis) for fusing the multi-modal
medical data. Specifically, we construct a multiplex network that incorporates
multiple types of non-image features of patients to capture the complex
relationship between patients in a systematic way, which leads to more accurate
clinical decisions. Extensive experiments on various real-world datasets
demonstrate the superiority and practicality of HetMed. The source code for
HetMed is available at https://github.com/Sein-Kim/Multimodal-Medical.
| [
{
"version": "v1",
"created": "Mon, 28 Nov 2022 09:14:36 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Feb 2023 02:38:12 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Mar 2023 04:49:40 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Kim",
"Sein",
""
],
[
"Lee",
"Namkyeong",
""
],
[
"Lee",
"Junseok",
""
],
[
"Hyun",
"Dongmin",
""
],
[
"Park",
"Chanyoung",
""
]
] | TITLE: Heterogeneous Graph Learning for Multi-modal Medical Data Analysis
ABSTRACT: Routine clinical visits of a patient produce not only image data, but also
non-image data containing clinical information regarding the patient, i.e.,
medical data is multi-modal in nature. Such heterogeneous modalities offer
different and complementary perspectives on the same patient, resulting in more
accurate clinical decisions when they are properly combined. However, despite
its significance, how to effectively fuse the multi-modal medical data into a
unified framework has received relatively little attention. In this paper, we
propose an effective graph-based framework called HetMed (Heterogeneous Graph
Learning for Multi-modal Medical Data Analysis) for fusing the multi-modal
medical data. Specifically, we construct a multiplex network that incorporates
multiple types of non-image features of patients to capture the complex
relationship between patients in a systematic way, which leads to more accurate
clinical decisions. Extensive experiments on various real-world datasets
demonstrate the superiority and practicality of HetMed. The source code for
HetMed is available at https://github.com/Sein-Kim/Multimodal-Medical.
| no_new_dataset | 0.950411 |
2212.12056 | Christopher Henry | M. Sokolov, J. L. Storie, C. J. Henry, C. D. Storie, J. Cameron, R. S.
{\O}deg{\aa}rd, V. Zubinaite, S. Stikbakke | Semantically-consistent Landsat 8 image to Sentinel-2 image translation
for alpine areas | 13 pages, 6 figures | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | The availability of frequent and cost-free satellite images is in growing
demand in the research world. Such satellite constellations as Landsat 8 and
Sentinel-2 provide a massive amount of valuable data daily. However, the
discrepancy in the sensors' characteristics of these satellites makes it
senseless to use a segmentation model trained on either dataset and applied to
another, which is why domain adaptation techniques have recently become an
active research area in remote sensing. In this paper, an experiment of domain
adaptation through style-transferring is conducted using the HRSemI2I model to
narrow the sensor discrepancy between Landsat 8 and Sentinel-2. This paper's
main contribution is analyzing the expediency of that approach by comparing the
results of segmentation using domain-adapted images with those without
adaptation. The HRSemI2I model, adjusted to work with 6-band imagery, shows
significant intersection-over-union performance improvement for both mean and
per class metrics. A second contribution is providing different schemes of
generalization between two label schemes - NALCMS 2015 and CORINE. The first
scheme is standardization through higher-level land cover classes, and the
second is through harmonization validation in the field.
| [
{
"version": "v1",
"created": "Thu, 22 Dec 2022 22:07:28 GMT"
}
] | 2023-03-16T00:00:00 | [
[
"Sokolov",
"M.",
""
],
[
"Storie",
"J. L.",
""
],
[
"Henry",
"C. J.",
""
],
[
"Storie",
"C. D.",
""
],
[
"Cameron",
"J.",
""
],
[
"Ødegård",
"R. S.",
""
],
[
"Zubinaite",
"V.",
""
],
[
"Stikbakke",
"S.",
""
]
] | TITLE: Semantically-consistent Landsat 8 image to Sentinel-2 image translation
for alpine areas
ABSTRACT: The availability of frequent and cost-free satellite images is in growing
demand in the research world. Such satellite constellations as Landsat 8 and
Sentinel-2 provide a massive amount of valuable data daily. However, the
discrepancy in the sensors' characteristics of these satellites makes it
senseless to use a segmentation model trained on either dataset and applied to
another, which is why domain adaptation techniques have recently become an
active research area in remote sensing. In this paper, an experiment of domain
adaptation through style-transferring is conducted using the HRSemI2I model to
narrow the sensor discrepancy between Landsat 8 and Sentinel-2. This paper's
main contribution is analyzing the expediency of that approach by comparing the
results of segmentation using domain-adapted images with those without
adaptation. The HRSemI2I model, adjusted to work with 6-band imagery, shows
significant intersection-over-union performance improvement for both mean and
per class metrics. A second contribution is providing different schemes of
generalization between two label schemes - NALCMS 2015 and CORINE. The first
scheme is standardization through higher-level land cover classes, and the
second is through harmonization validation in the field.
| no_new_dataset | 0.948394 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.