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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1306.0239 | Yichuan Tang | Yichuan Tang | Deep Learning using Linear Support Vector Machines | Contribution to the ICML 2013 Challenges in Representation Learning
Workshop | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, fully-connected and convolutional neural networks have been trained
to achieve state-of-the-art performance on a wide variety of tasks such as
speech recognition, image classification, natural language processing, and
bioinformatics. For classification tasks, most of these "deep learning" models
employ the softmax activation function for prediction and minimize
cross-entropy loss. In this paper, we demonstrate a small but consistent
advantage of replacing the softmax layer with a linear support vector machine.
Learning minimizes a margin-based loss instead of the cross-entropy loss. While
there have been various combinations of neural nets and SVMs in prior art, our
results using L2-SVMs show that by simply replacing softmax with linear SVMs
gives significant gains on popular deep learning datasets MNIST, CIFAR-10, and
the ICML 2013 Representation Learning Workshop's face expression recognition
challenge.
| [
{
"version": "v1",
"created": "Sun, 2 Jun 2013 18:46:58 GMT"
},
{
"version": "v2",
"created": "Tue, 9 Jul 2013 21:30:59 GMT"
},
{
"version": "v3",
"created": "Mon, 23 Dec 2013 21:16:45 GMT"
},
{
"version": "v4",
"created": "Sat, 21 Feb 2015 16:58:39 GMT"
}
] | 2015-02-24T00:00:00 | [
[
"Tang",
"Yichuan",
""
]
] | TITLE: Deep Learning using Linear Support Vector Machines
ABSTRACT: Recently, fully-connected and convolutional neural networks have been trained
to achieve state-of-the-art performance on a wide variety of tasks such as
speech recognition, image classification, natural language processing, and
bioinformatics. For classification tasks, most of these "deep learning" models
employ the softmax activation function for prediction and minimize
cross-entropy loss. In this paper, we demonstrate a small but consistent
advantage of replacing the softmax layer with a linear support vector machine.
Learning minimizes a margin-based loss instead of the cross-entropy loss. While
there have been various combinations of neural nets and SVMs in prior art, our
results using L2-SVMs show that by simply replacing softmax with linear SVMs
gives significant gains on popular deep learning datasets MNIST, CIFAR-10, and
the ICML 2013 Representation Learning Workshop's face expression recognition
challenge.
|
1312.6110 | Yichuan Tang | Yichuan Tang, Nitish Srivastava, Ruslan Salakhutdinov | Learning Generative Models with Visual Attention | In the proceedings of Neural Information Processing Systems, 2014 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Attention has long been proposed by psychologists as important for
effectively dealing with the enormous sensory stimulus available in the
neocortex. Inspired by the visual attention models in computational
neuroscience and the need of object-centric data for generative models, we
describe for generative learning framework using attentional mechanisms.
Attentional mechanisms can propagate signals from region of interest in a scene
to an aligned canonical representation, where generative modeling takes place.
By ignoring background clutter, generative models can concentrate their
resources on the object of interest. Our model is a proper graphical model
where the 2D Similarity transformation is a part of the top-down process. A
ConvNet is employed to provide good initializations during posterior inference
which is based on Hamiltonian Monte Carlo. Upon learning images of faces, our
model can robustly attend to face regions of novel test subjects. More
importantly, our model can learn generative models of new faces from a novel
dataset of large images where the face locations are not known.
| [
{
"version": "v1",
"created": "Fri, 20 Dec 2013 20:50:43 GMT"
},
{
"version": "v2",
"created": "Mon, 30 Dec 2013 16:49:43 GMT"
},
{
"version": "v3",
"created": "Sat, 21 Feb 2015 22:21:15 GMT"
}
] | 2015-02-24T00:00:00 | [
[
"Tang",
"Yichuan",
""
],
[
"Srivastava",
"Nitish",
""
],
[
"Salakhutdinov",
"Ruslan",
""
]
] | TITLE: Learning Generative Models with Visual Attention
ABSTRACT: Attention has long been proposed by psychologists as important for
effectively dealing with the enormous sensory stimulus available in the
neocortex. Inspired by the visual attention models in computational
neuroscience and the need of object-centric data for generative models, we
describe for generative learning framework using attentional mechanisms.
Attentional mechanisms can propagate signals from region of interest in a scene
to an aligned canonical representation, where generative modeling takes place.
By ignoring background clutter, generative models can concentrate their
resources on the object of interest. Our model is a proper graphical model
where the 2D Similarity transformation is a part of the top-down process. A
ConvNet is employed to provide good initializations during posterior inference
which is based on Hamiltonian Monte Carlo. Upon learning images of faces, our
model can robustly attend to face regions of novel test subjects. More
importantly, our model can learn generative models of new faces from a novel
dataset of large images where the face locations are not known.
|
1405.0312 | Piotr Doll\'ar | Tsung-Yi Lin, Michael Maire, Serge Belongie, Lubomir Bourdev, Ross
Girshick, James Hays, Pietro Perona, Deva Ramanan, C. Lawrence Zitnick, Piotr
Doll\'ar | Microsoft COCO: Common Objects in Context | 1) updated annotation pipeline description and figures; 2) added new
section describing datasets splits; 3) updated author list | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a new dataset with the goal of advancing the state-of-the-art in
object recognition by placing the question of object recognition in the context
of the broader question of scene understanding. This is achieved by gathering
images of complex everyday scenes containing common objects in their natural
context. Objects are labeled using per-instance segmentations to aid in precise
object localization. Our dataset contains photos of 91 objects types that would
be easily recognizable by a 4 year old. With a total of 2.5 million labeled
instances in 328k images, the creation of our dataset drew upon extensive crowd
worker involvement via novel user interfaces for category detection, instance
spotting and instance segmentation. We present a detailed statistical analysis
of the dataset in comparison to PASCAL, ImageNet, and SUN. Finally, we provide
baseline performance analysis for bounding box and segmentation detection
results using a Deformable Parts Model.
| [
{
"version": "v1",
"created": "Thu, 1 May 2014 21:43:32 GMT"
},
{
"version": "v2",
"created": "Sat, 5 Jul 2014 18:39:56 GMT"
},
{
"version": "v3",
"created": "Sat, 21 Feb 2015 01:48:49 GMT"
}
] | 2015-02-24T00:00:00 | [
[
"Lin",
"Tsung-Yi",
""
],
[
"Maire",
"Michael",
""
],
[
"Belongie",
"Serge",
""
],
[
"Bourdev",
"Lubomir",
""
],
[
"Girshick",
"Ross",
""
],
[
"Hays",
"James",
""
],
[
"Perona",
"Pietro",
""
],
[
"Ramanan",
"Deva",
""
],
[
"Zitnick",
"C. Lawrence",
""
],
[
"Dollár",
"Piotr",
""
]
] | TITLE: Microsoft COCO: Common Objects in Context
ABSTRACT: We present a new dataset with the goal of advancing the state-of-the-art in
object recognition by placing the question of object recognition in the context
of the broader question of scene understanding. This is achieved by gathering
images of complex everyday scenes containing common objects in their natural
context. Objects are labeled using per-instance segmentations to aid in precise
object localization. Our dataset contains photos of 91 objects types that would
be easily recognizable by a 4 year old. With a total of 2.5 million labeled
instances in 328k images, the creation of our dataset drew upon extensive crowd
worker involvement via novel user interfaces for category detection, instance
spotting and instance segmentation. We present a detailed statistical analysis
of the dataset in comparison to PASCAL, ImageNet, and SUN. Finally, we provide
baseline performance analysis for bounding box and segmentation detection
results using a Deformable Parts Model.
|
1412.6039 | Xin Yuan | Yunchen Pu, Xin Yuan and Lawrence Carin | Generative Deep Deconvolutional Learning | 21 pages, 9 figures, revised version for ICLR 2015 | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A generative Bayesian model is developed for deep (multi-layer) convolutional
dictionary learning. A novel probabilistic pooling operation is integrated into
the deep model, yielding efficient bottom-up and top-down probabilistic
learning. After learning the deep convolutional dictionary, testing is
implemented via deconvolutional inference. To speed up this inference, a new
statistical approach is proposed to project the top-layer dictionary elements
to the data level. Following this, only one layer of deconvolution is required
during testing. Experimental results demonstrate powerful capabilities of the
model to learn multi-layer features from images. Excellent classification
results are obtained on both the MNIST and Caltech 101 datasets.
| [
{
"version": "v1",
"created": "Thu, 18 Dec 2014 20:01:38 GMT"
},
{
"version": "v2",
"created": "Fri, 19 Dec 2014 17:21:36 GMT"
},
{
"version": "v3",
"created": "Sun, 22 Feb 2015 18:13:29 GMT"
}
] | 2015-02-24T00:00:00 | [
[
"Pu",
"Yunchen",
""
],
[
"Yuan",
"Xin",
""
],
[
"Carin",
"Lawrence",
""
]
] | TITLE: Generative Deep Deconvolutional Learning
ABSTRACT: A generative Bayesian model is developed for deep (multi-layer) convolutional
dictionary learning. A novel probabilistic pooling operation is integrated into
the deep model, yielding efficient bottom-up and top-down probabilistic
learning. After learning the deep convolutional dictionary, testing is
implemented via deconvolutional inference. To speed up this inference, a new
statistical approach is proposed to project the top-layer dictionary elements
to the data level. Following this, only one layer of deconvolution is required
during testing. Experimental results demonstrate powerful capabilities of the
model to learn multi-layer features from images. Excellent classification
results are obtained on both the MNIST and Caltech 101 datasets.
|
1502.06064 | Ken Miura | Ken Miura, Tetsuaki Mano, Atsushi Kanehira, Yuichiro Tsuchiya and
Tatsuya Harada | MILJS : Brand New JavaScript Libraries for Matrix Calculation and
Machine Learning | null | null | null | null | stat.ML cs.LG cs.MS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | MILJS is a collection of state-of-the-art, platform-independent, scalable,
fast JavaScript libraries for matrix calculation and machine learning. Our core
library offering a matrix calculation is called Sushi, which exhibits far
better performance than any other leading machine learning libraries written in
JavaScript. Especially, our matrix multiplication is 177 times faster than the
fastest JavaScript benchmark. Based on Sushi, a machine learning library called
Tempura is provided, which supports various algorithms widely used in machine
learning research. We also provide Soba as a visualization library. The
implementations of our libraries are clearly written, properly documented and
thus can are easy to get started with, as long as there is a web browser. These
libraries are available from http://mil-tokyo.github.io/ under the MIT license.
| [
{
"version": "v1",
"created": "Sat, 21 Feb 2015 04:29:41 GMT"
}
] | 2015-02-24T00:00:00 | [
[
"Miura",
"Ken",
""
],
[
"Mano",
"Tetsuaki",
""
],
[
"Kanehira",
"Atsushi",
""
],
[
"Tsuchiya",
"Yuichiro",
""
],
[
"Harada",
"Tatsuya",
""
]
] | TITLE: MILJS : Brand New JavaScript Libraries for Matrix Calculation and
Machine Learning
ABSTRACT: MILJS is a collection of state-of-the-art, platform-independent, scalable,
fast JavaScript libraries for matrix calculation and machine learning. Our core
library offering a matrix calculation is called Sushi, which exhibits far
better performance than any other leading machine learning libraries written in
JavaScript. Especially, our matrix multiplication is 177 times faster than the
fastest JavaScript benchmark. Based on Sushi, a machine learning library called
Tempura is provided, which supports various algorithms widely used in machine
learning research. We also provide Soba as a visualization library. The
implementations of our libraries are clearly written, properly documented and
thus can are easy to get started with, as long as there is a web browser. These
libraries are available from http://mil-tokyo.github.io/ under the MIT license.
|
1502.06086 | Suyash Gupta | Suyash Gupta, Rahul Shrivastava and V. Krishna Nandivada | DCAFE: Dynamic load-balanced loop Chunking & Aggressive Finish
Elimination for Recursive Task Parallel Programs | null | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we present two symbiotic optimizations to optimize recursive
task parallel (RTP) programs by reducing the task creation and termination
overheads. Our first optimization Aggressive Finish-Elimination (AFE) helps
reduce the redundant join operations to a large extent. The second optimization
Dynamic Load-Balanced loop Chunking (DLBC) extends the prior work on loop
chunking to decide on the number of parallel tasks based on the number of
available worker threads, at runtime. Further, we discuss the impact of
exceptions on our optimizations and extend them to handle RTP programs that may
throw exceptions. We implemented DCAFE (= DLBC+AFE) in the X10v2.3 compiler and
tested it over a set of benchmark kernels on two different hardwares (a 16-core
Intel system and a 64-core AMD system). With respect to the base X10 compiler
extended with loop-chunking of Nandivada et al [Nandivada et
al.(2013)Nandivada, Shirako, Zhao, and Sarkar](LC), DCAFE achieved a geometric
mean speed up of 5.75x and 4.16x on the Intel and AMD system, respectively. We
also present an evaluation with respect to the energy consumption on the Intel
system and show that on average, compared to the LC versions, the DCAFE
versions consume 71.2% less energy.
| [
{
"version": "v1",
"created": "Sat, 21 Feb 2015 08:18:43 GMT"
}
] | 2015-02-24T00:00:00 | [
[
"Gupta",
"Suyash",
""
],
[
"Shrivastava",
"Rahul",
""
],
[
"Nandivada",
"V. Krishna",
""
]
] | TITLE: DCAFE: Dynamic load-balanced loop Chunking & Aggressive Finish
Elimination for Recursive Task Parallel Programs
ABSTRACT: In this paper, we present two symbiotic optimizations to optimize recursive
task parallel (RTP) programs by reducing the task creation and termination
overheads. Our first optimization Aggressive Finish-Elimination (AFE) helps
reduce the redundant join operations to a large extent. The second optimization
Dynamic Load-Balanced loop Chunking (DLBC) extends the prior work on loop
chunking to decide on the number of parallel tasks based on the number of
available worker threads, at runtime. Further, we discuss the impact of
exceptions on our optimizations and extend them to handle RTP programs that may
throw exceptions. We implemented DCAFE (= DLBC+AFE) in the X10v2.3 compiler and
tested it over a set of benchmark kernels on two different hardwares (a 16-core
Intel system and a 64-core AMD system). With respect to the base X10 compiler
extended with loop-chunking of Nandivada et al [Nandivada et
al.(2013)Nandivada, Shirako, Zhao, and Sarkar](LC), DCAFE achieved a geometric
mean speed up of 5.75x and 4.16x on the Intel and AMD system, respectively. We
also present an evaluation with respect to the energy consumption on the Intel
system and show that on average, compared to the LC versions, the DCAFE
versions consume 71.2% less energy.
|
1502.06219 | Smitha M.L. | B.H. Shekar, Smitha M.L. | Video Text Localization with an emphasis on Edge Features | 8 pages, Eighth International Conference on Image and Signal
Processing, Elsevier Publications, ISBN: 9789351072522, pp: 324-330, held at
UVCE, Bangalore in July 2014. arXiv admin note: text overlap with
arXiv:1502.03913 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The text detection and localization plays a major role in video analysis and
understanding. The scene text embedded in video consist of high-level semantics
and hence contributes significantly to visual content analysis and retrieval.
This paper proposes a novel method to robustly localize the texts in natural
scene images and videos based on sobel edge emphasizing approach. The input
image is preprocessed and edge emphasis is done to detect the text clusters.
Further, a set of rules have been devised using morphological operators for
false positive elimination and connected component analysis is performed to
detect the text regions and hence text localization is performed. The
experimental results obtained on publicly available standard datasets
illustrate that the proposed method can detect and localize the texts of
various sizes, fonts and colors.
| [
{
"version": "v1",
"created": "Sun, 22 Feb 2015 12:32:18 GMT"
}
] | 2015-02-24T00:00:00 | [
[
"Shekar",
"B. H.",
""
],
[
"L.",
"Smitha M.",
""
]
] | TITLE: Video Text Localization with an emphasis on Edge Features
ABSTRACT: The text detection and localization plays a major role in video analysis and
understanding. The scene text embedded in video consist of high-level semantics
and hence contributes significantly to visual content analysis and retrieval.
This paper proposes a novel method to robustly localize the texts in natural
scene images and videos based on sobel edge emphasizing approach. The input
image is preprocessed and edge emphasis is done to detect the text clusters.
Further, a set of rules have been devised using morphological operators for
false positive elimination and connected component analysis is performed to
detect the text regions and hence text localization is performed. The
experimental results obtained on publicly available standard datasets
illustrate that the proposed method can detect and localize the texts of
various sizes, fonts and colors.
|
1502.06235 | Borislav Antic | Borislav Anti\'c and Bj\"orn Ommer | Spatio-temporal Video Parsing for Abnormality Detection | 15 pages, 12 figures, 3 tables | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Abnormality detection in video poses particular challenges due to the
infinite size of the class of all irregular objects and behaviors. Thus no (or
by far not enough) abnormal training samples are available and we need to find
abnormalities in test data without actually knowing what they are.
Nevertheless, the prevailing concept of the field is to directly search for
individual abnormal local patches or image regions independent of another. To
address this problem, we propose a method for joint detection of abnormalities
in videos by spatio-temporal video parsing. The goal of video parsing is to
find a set of indispensable normal spatio-temporal object hypotheses that
jointly explain all the foreground of a video, while, at the same time, being
supported by normal training samples. Consequently, we avoid a direct detection
of abnormalities and discover them indirectly as those hypotheses which are
needed for covering the foreground without finding an explanation for
themselves by normal samples. Abnormalities are localized by MAP inference in a
graphical model and we solve it efficiently by formulating it as a convex
optimization problem. We experimentally evaluate our approach on several
challenging benchmark sets, improving over the state-of-the-art on all standard
benchmarks both in terms of abnormality classification and localization.
| [
{
"version": "v1",
"created": "Sun, 22 Feb 2015 14:54:02 GMT"
}
] | 2015-02-24T00:00:00 | [
[
"Antić",
"Borislav",
""
],
[
"Ommer",
"Björn",
""
]
] | TITLE: Spatio-temporal Video Parsing for Abnormality Detection
ABSTRACT: Abnormality detection in video poses particular challenges due to the
infinite size of the class of all irregular objects and behaviors. Thus no (or
by far not enough) abnormal training samples are available and we need to find
abnormalities in test data without actually knowing what they are.
Nevertheless, the prevailing concept of the field is to directly search for
individual abnormal local patches or image regions independent of another. To
address this problem, we propose a method for joint detection of abnormalities
in videos by spatio-temporal video parsing. The goal of video parsing is to
find a set of indispensable normal spatio-temporal object hypotheses that
jointly explain all the foreground of a video, while, at the same time, being
supported by normal training samples. Consequently, we avoid a direct detection
of abnormalities and discover them indirectly as those hypotheses which are
needed for covering the foreground without finding an explanation for
themselves by normal samples. Abnormalities are localized by MAP inference in a
graphical model and we solve it efficiently by formulating it as a convex
optimization problem. We experimentally evaluate our approach on several
challenging benchmark sets, improving over the state-of-the-art on all standard
benchmarks both in terms of abnormality classification and localization.
|
1502.06344 | Erik Rodner | Clemens-Alexander Brust, Sven Sickert, Marcel Simon, Erik Rodner,
Joachim Denzler | Convolutional Patch Networks with Spatial Prior for Road Detection and
Urban Scene Understanding | VISAPP 2015 paper | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Classifying single image patches is important in many different applications,
such as road detection or scene understanding. In this paper, we present
convolutional patch networks, which are convolutional networks learned to
distinguish different image patches and which can be used for pixel-wise
labeling. We also show how to incorporate spatial information of the patch as
an input to the network, which allows for learning spatial priors for certain
categories jointly with an appearance model. In particular, we focus on road
detection and urban scene understanding, two application areas where we are
able to achieve state-of-the-art results on the KITTI as well as on the
LabelMeFacade dataset.
Furthermore, our paper offers a guideline for people working in the area and
desperately wandering through all the painstaking details that render training
CNs on image patches extremely difficult.
| [
{
"version": "v1",
"created": "Mon, 23 Feb 2015 08:47:37 GMT"
}
] | 2015-02-24T00:00:00 | [
[
"Brust",
"Clemens-Alexander",
""
],
[
"Sickert",
"Sven",
""
],
[
"Simon",
"Marcel",
""
],
[
"Rodner",
"Erik",
""
],
[
"Denzler",
"Joachim",
""
]
] | TITLE: Convolutional Patch Networks with Spatial Prior for Road Detection and
Urban Scene Understanding
ABSTRACT: Classifying single image patches is important in many different applications,
such as road detection or scene understanding. In this paper, we present
convolutional patch networks, which are convolutional networks learned to
distinguish different image patches and which can be used for pixel-wise
labeling. We also show how to incorporate spatial information of the patch as
an input to the network, which allows for learning spatial priors for certain
categories jointly with an appearance model. In particular, we focus on road
detection and urban scene understanding, two application areas where we are
able to achieve state-of-the-art results on the KITTI as well as on the
LabelMeFacade dataset.
Furthermore, our paper offers a guideline for people working in the area and
desperately wandering through all the painstaking details that render training
CNs on image patches extremely difficult.
|
1502.03913 | Smitha M.L. | B.H. Shekar, Smitha M.L | Skeleton Matching based approach for Text Localization in Scene Images | 10 pages, 8 figures, Eighth International Conference on Image and
Signal Processing,Elsevier Publications,pp: 145-153, held at UVCE, Bangalore
in July 2014. ISBN: 9789351072522 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose a skeleton matching based approach which aids in
text localization in scene images. The input image is preprocessed and
segmented into blocks using connected component analysis. We obtain the
skeleton of the segmented block using morphology based approach. The
skeletonized images are compared with the trained templates in the database to
categorize into text and non-text blocks. Further, the newly designed
geometrical rules and morphological operations are employed on the detected
text blocks for scene text localization. The experimental results obtained on
publicly available standard datasets illustrate that the proposed method can
detect and localize the texts of various sizes, fonts and colors.
| [
{
"version": "v1",
"created": "Fri, 13 Feb 2015 08:42:00 GMT"
}
] | 2015-02-23T00:00:00 | [
[
"Shekar",
"B. H.",
""
],
[
"L",
"Smitha M.",
""
]
] | TITLE: Skeleton Matching based approach for Text Localization in Scene Images
ABSTRACT: In this paper, we propose a skeleton matching based approach which aids in
text localization in scene images. The input image is preprocessed and
segmented into blocks using connected component analysis. We obtain the
skeleton of the segmented block using morphology based approach. The
skeletonized images are compared with the trained templates in the database to
categorize into text and non-text blocks. Further, the newly designed
geometrical rules and morphological operations are employed on the detected
text blocks for scene text localization. The experimental results obtained on
publicly available standard datasets illustrate that the proposed method can
detect and localize the texts of various sizes, fonts and colors.
|
1502.03918 | Smitha M.L. | B.H. Shekar, Smitha M.L | Gradient Difference based approach for Text Localization in Compressed
domain | 11 pages, Second International Conference on Emerging Research in
Computing, Information, Communications and Applications, Elsevier
Publications, ISBN: 9789351072638, vol. III, pp: 299-308, held at NMIT,
Bangalore August 2014 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose a gradient difference based approach to text
localization in videos and scene images. The input video frame/ image is first
compressed using multilevel 2-D wavelet transform. The edge information of the
reconstructed image is found which is further used for finding the maximum
gradient difference between the pixels and then the boundaries of the detected
text blocks are computed using zero crossing technique. We perform logical AND
operation of the text blocks obtained by gradient difference and the zero
crossing technique followed by connected component analysis to eliminate the
false positives. Finally, the morphological dilation operation is employed on
the detected text blocks for scene text localization. The experimental results
obtained on publicly available standard datasets illustrate that the proposed
method can detect and localize the texts of various sizes, fonts and colors.
| [
{
"version": "v1",
"created": "Fri, 13 Feb 2015 09:08:35 GMT"
}
] | 2015-02-23T00:00:00 | [
[
"Shekar",
"B. H.",
""
],
[
"L",
"Smitha M.",
""
]
] | TITLE: Gradient Difference based approach for Text Localization in Compressed
domain
ABSTRACT: In this paper, we propose a gradient difference based approach to text
localization in videos and scene images. The input video frame/ image is first
compressed using multilevel 2-D wavelet transform. The edge information of the
reconstructed image is found which is further used for finding the maximum
gradient difference between the pixels and then the boundaries of the detected
text blocks are computed using zero crossing technique. We perform logical AND
operation of the text blocks obtained by gradient difference and the zero
crossing technique followed by connected component analysis to eliminate the
false positives. Finally, the morphological dilation operation is employed on
the detected text blocks for scene text localization. The experimental results
obtained on publicly available standard datasets illustrate that the proposed
method can detect and localize the texts of various sizes, fonts and colors.
|
1502.05925 | Feng Nan | Feng Nan, Joseph Wang, Venkatesh Saligrama | Feature-Budgeted Random Forest | null | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We seek decision rules for prediction-time cost reduction, where complete
data is available for training, but during prediction-time, each feature can
only be acquired for an additional cost. We propose a novel random forest
algorithm to minimize prediction error for a user-specified {\it average}
feature acquisition budget. While random forests yield strong generalization
performance, they do not explicitly account for feature costs and furthermore
require low correlation among trees, which amplifies costs. Our random forest
grows trees with low acquisition cost and high strength based on greedy minimax
cost-weighted-impurity splits. Theoretically, we establish near-optimal
acquisition cost guarantees for our algorithm. Empirically, on a number of
benchmark datasets we demonstrate superior accuracy-cost curves against
state-of-the-art prediction-time algorithms.
| [
{
"version": "v1",
"created": "Fri, 20 Feb 2015 16:42:40 GMT"
}
] | 2015-02-23T00:00:00 | [
[
"Nan",
"Feng",
""
],
[
"Wang",
"Joseph",
""
],
[
"Saligrama",
"Venkatesh",
""
]
] | TITLE: Feature-Budgeted Random Forest
ABSTRACT: We seek decision rules for prediction-time cost reduction, where complete
data is available for training, but during prediction-time, each feature can
only be acquired for an additional cost. We propose a novel random forest
algorithm to minimize prediction error for a user-specified {\it average}
feature acquisition budget. While random forests yield strong generalization
performance, they do not explicitly account for feature costs and furthermore
require low correlation among trees, which amplifies costs. Our random forest
grows trees with low acquisition cost and high strength based on greedy minimax
cost-weighted-impurity splits. Theoretically, we establish near-optimal
acquisition cost guarantees for our algorithm. Empirically, on a number of
benchmark datasets we demonstrate superior accuracy-cost curves against
state-of-the-art prediction-time algorithms.
|
1502.04272 | Alex James Dr | Joshin John Mathew and Alex Pappachen James | Spatial Stimuli Gradient Sketch Model | accepted for publication in IEEE Signal Processing Letters, 2015 | Volume: 22 Issue: 9 On page(s): 1336-1339, 2015 | 10.1109/LSP.2015.2404827 | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/3.0/ | The inability of automated edge detection methods inspired from primal sketch
models to accurately calculate object edges under the influence of pixel noise
is an open problem. Extending the principles of image perception i.e.
Weber-Fechner law, and Sheperd similarity law, we propose a new edge detection
method and formulation that use perceived brightness and neighbourhood
similarity calculations in the determination of robust object edges. The
robustness of the detected edges is benchmark against Sobel, SIS, Kirsch, and
Prewitt edge detection methods in an example face recognition problem showing
statistically significant improvement in recognition accuracy and pixel noise
tolerance.
| [
{
"version": "v1",
"created": "Sun, 15 Feb 2015 02:42:56 GMT"
}
] | 2015-02-20T00:00:00 | [
[
"Mathew",
"Joshin John",
""
],
[
"James",
"Alex Pappachen",
""
]
] | TITLE: Spatial Stimuli Gradient Sketch Model
ABSTRACT: The inability of automated edge detection methods inspired from primal sketch
models to accurately calculate object edges under the influence of pixel noise
is an open problem. Extending the principles of image perception i.e.
Weber-Fechner law, and Sheperd similarity law, we propose a new edge detection
method and formulation that use perceived brightness and neighbourhood
similarity calculations in the determination of robust object edges. The
robustness of the detected edges is benchmark against Sobel, SIS, Kirsch, and
Prewitt edge detection methods in an example face recognition problem showing
statistically significant improvement in recognition accuracy and pixel noise
tolerance.
|
1502.05461 | Carl Vondrick | Carl Vondrick, Aditya Khosla, Hamed Pirsiavash, Tomasz Malisiewicz,
Antonio Torralba | Visualizing Object Detection Features | In submission to IJCV | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce algorithms to visualize feature spaces used by object detectors.
Our method works by inverting a visual feature back to multiple natural images.
We found that these visualizations allow us to analyze object detection systems
in new ways and gain new insight into the detector's failures. For example,
when we visualize the features for high scoring false alarms, we discovered
that, although they are clearly wrong in image space, they do look deceptively
similar to true positives in feature space. This result suggests that many of
these false alarms are caused by our choice of feature space, and supports that
creating a better learning algorithm or building bigger datasets is unlikely to
correct these errors. By visualizing feature spaces, we can gain a more
intuitive understanding of recognition systems.
| [
{
"version": "v1",
"created": "Thu, 19 Feb 2015 04:11:14 GMT"
}
] | 2015-02-20T00:00:00 | [
[
"Vondrick",
"Carl",
""
],
[
"Khosla",
"Aditya",
""
],
[
"Pirsiavash",
"Hamed",
""
],
[
"Malisiewicz",
"Tomasz",
""
],
[
"Torralba",
"Antonio",
""
]
] | TITLE: Visualizing Object Detection Features
ABSTRACT: We introduce algorithms to visualize feature spaces used by object detectors.
Our method works by inverting a visual feature back to multiple natural images.
We found that these visualizations allow us to analyze object detection systems
in new ways and gain new insight into the detector's failures. For example,
when we visualize the features for high scoring false alarms, we discovered
that, although they are clearly wrong in image space, they do look deceptively
similar to true positives in feature space. This result suggests that many of
these false alarms are caused by our choice of feature space, and supports that
creating a better learning algorithm or building bigger datasets is unlikely to
correct these errors. By visualizing feature spaces, we can gain a more
intuitive understanding of recognition systems.
|
1502.04868 | Rafael Boloix-Tortosa | Rafael Boloix-Tortosa, F. Javier Pay\'an-Somet, Eva Arias-de-Reyna and
Juan Jos\'e Murillo-Fuentes | Proper Complex Gaussian Processes for Regression | null | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Complex-valued signals are used in the modeling of many systems in
engineering and science, hence being of fundamental interest. Often, random
complex-valued signals are considered to be proper. A proper complex random
variable or process is uncorrelated with its complex conjugate. This assumption
is a good model of the underlying physics in many problems, and simplifies the
computations. While linear processing and neural networks have been widely
studied for these signals, the development of complex-valued nonlinear kernel
approaches remains an open problem. In this paper we propose Gaussian processes
for regression as a framework to develop 1) a solution for proper
complex-valued kernel regression and 2) the design of the reproducing kernel
for complex-valued inputs, using the convolutional approach for
cross-covariances. In this design we pay attention to preserve, in the complex
domain, the measure of similarity between near inputs. The hyperparameters of
the kernel are learned maximizing the marginal likelihood using Wirtinger
derivatives. Besides, the approach is connected to the multiple output learning
scenario. In the experiments included, we first solve a proper complex Gaussian
process where the cross-covariance does not cancel, a challenging scenario when
dealing with proper complex signals. Then we successfully use these novel
results to solve some problems previously proposed in the literature as
benchmarks, reporting a remarkable improvement in the estimation error.
| [
{
"version": "v1",
"created": "Tue, 17 Feb 2015 11:59:44 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Feb 2015 09:33:34 GMT"
}
] | 2015-02-19T00:00:00 | [
[
"Boloix-Tortosa",
"Rafael",
""
],
[
"Payán-Somet",
"F. Javier",
""
],
[
"Arias-de-Reyna",
"Eva",
""
],
[
"Murillo-Fuentes",
"Juan José",
""
]
] | TITLE: Proper Complex Gaussian Processes for Regression
ABSTRACT: Complex-valued signals are used in the modeling of many systems in
engineering and science, hence being of fundamental interest. Often, random
complex-valued signals are considered to be proper. A proper complex random
variable or process is uncorrelated with its complex conjugate. This assumption
is a good model of the underlying physics in many problems, and simplifies the
computations. While linear processing and neural networks have been widely
studied for these signals, the development of complex-valued nonlinear kernel
approaches remains an open problem. In this paper we propose Gaussian processes
for regression as a framework to develop 1) a solution for proper
complex-valued kernel regression and 2) the design of the reproducing kernel
for complex-valued inputs, using the convolutional approach for
cross-covariances. In this design we pay attention to preserve, in the complex
domain, the measure of similarity between near inputs. The hyperparameters of
the kernel are learned maximizing the marginal likelihood using Wirtinger
derivatives. Besides, the approach is connected to the multiple output learning
scenario. In the experiments included, we first solve a proper complex Gaussian
process where the cross-covariance does not cancel, a challenging scenario when
dealing with proper complex signals. Then we successfully use these novel
results to solve some problems previously proposed in the literature as
benchmarks, reporting a remarkable improvement in the estimation error.
|
1502.05167 | Mansaf Alam Dr | Kashish Ara Shakil, Shadma Anis and Mansaf Alam | Dengue disease prediction using weka data mining tool | null | null | null | null | cs.CY cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Dengue is a life threatening disease prevalent in several developed as well
as developing countries like India.In this paper we discuss various algorithm
approaches of data mining that have been utilized for dengue disease
prediction. Data mining is a well known technique used by health organizations
for classification of diseases such as dengue, diabetes and cancer in
bioinformatics research. In the proposed approach we have used WEKA with 10
cross validation to evaluate data and compare results. Weka has an extensive
collection of different machine learning and data mining algorithms. In this
paper we have firstly classified the dengue data set and then compared the
different data mining techniques in weka through Explorer, knowledge flow and
Experimenter interfaces. Furthermore in order to validate our approach we have
used a dengue dataset with 108 instances but weka used 99 rows and 18
attributes to determine the prediction of disease and their accuracy using
classifications of different algorithms to find out the best performance. The
main objective of this paper is to classify data and assist the users in
extracting useful information from data and easily identify a suitable
algorithm for accurate predictive model from it. From the findings of this
paper it can be concluded that Na\"ive Bayes and J48 are the best performance
algorithms for classified accuracy because they achieved maximum accuracy= 100%
with 99 correctly classified instances, maximum ROC = 1, had least mean
absolute error and it took minimum time for building this model through
Explorer and Knowledge flow results
| [
{
"version": "v1",
"created": "Wed, 18 Feb 2015 09:52:55 GMT"
}
] | 2015-02-19T00:00:00 | [
[
"Shakil",
"Kashish Ara",
""
],
[
"Anis",
"Shadma",
""
],
[
"Alam",
"Mansaf",
""
]
] | TITLE: Dengue disease prediction using weka data mining tool
ABSTRACT: Dengue is a life threatening disease prevalent in several developed as well
as developing countries like India.In this paper we discuss various algorithm
approaches of data mining that have been utilized for dengue disease
prediction. Data mining is a well known technique used by health organizations
for classification of diseases such as dengue, diabetes and cancer in
bioinformatics research. In the proposed approach we have used WEKA with 10
cross validation to evaluate data and compare results. Weka has an extensive
collection of different machine learning and data mining algorithms. In this
paper we have firstly classified the dengue data set and then compared the
different data mining techniques in weka through Explorer, knowledge flow and
Experimenter interfaces. Furthermore in order to validate our approach we have
used a dengue dataset with 108 instances but weka used 99 rows and 18
attributes to determine the prediction of disease and their accuracy using
classifications of different algorithms to find out the best performance. The
main objective of this paper is to classify data and assist the users in
extracting useful information from data and easily identify a suitable
algorithm for accurate predictive model from it. From the findings of this
paper it can be concluded that Na\"ive Bayes and J48 are the best performance
algorithms for classified accuracy because they achieved maximum accuracy= 100%
with 99 correctly classified instances, maximum ROC = 1, had least mean
absolute error and it took minimum time for building this model through
Explorer and Knowledge flow results
|
1502.05212 | Gianluigi Cioca | Gianluigi Ciocca, Paolo Napoletano, Raimondo Schettini | IAT - Image Annotation Tool: Manual | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The annotation of image and video data of large datasets is a fundamental
task in multimedia information retrieval and computer vision applications. In
order to support the users during the image and video annotation process,
several software tools have been developed to provide them with a graphical
environment which helps drawing object contours, handling tracking information
and specifying object metadata. Here we introduce a preliminary version of the
image annotation tools developed at the Imaging and Vision Laboratory.
| [
{
"version": "v1",
"created": "Wed, 18 Feb 2015 13:11:46 GMT"
}
] | 2015-02-19T00:00:00 | [
[
"Ciocca",
"Gianluigi",
""
],
[
"Napoletano",
"Paolo",
""
],
[
"Schettini",
"Raimondo",
""
]
] | TITLE: IAT - Image Annotation Tool: Manual
ABSTRACT: The annotation of image and video data of large datasets is a fundamental
task in multimedia information retrieval and computer vision applications. In
order to support the users during the image and video annotation process,
several software tools have been developed to provide them with a graphical
environment which helps drawing object contours, handling tracking information
and specifying object metadata. Here we introduce a preliminary version of the
image annotation tools developed at the Imaging and Vision Laboratory.
|
1502.05241 | Michael Dirnberger | Michael Dirnberger and Adrian Neumann and Tim Kehl | NEFI: Network Extraction From Images | null | null | null | null | cs.CV cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Networks and network-like structures are amongst the central building blocks
of many technological and biological systems. Given a mathematical graph
representation of a network, methods from graph theory enable a precise
investigation of its properties. Software for the analysis of graphs is widely
available and has been applied to graphs describing large scale networks such
as social networks, protein-interaction networks, etc. In these applications,
graph acquisition, i.e., the extraction of a mathematical graph from a network,
is relatively simple. However, for many network-like structures, e.g. leaf
venations, slime molds and mud cracks, data collection relies on images where
graph extraction requires domain-specific solutions or even manual. Here we
introduce Network Extraction From Images, NEFI, a software tool that
automatically extracts accurate graphs from images of a wide range of networks
originating in various domains. While there is previous work on graph
extraction from images, theoretical results are fully accessible only to an
expert audience and ready-to-use implementations for non-experts are rarely
available or insufficiently documented. NEFI provides a novel platform allowing
practitioners from many disciplines to easily extract graph representations
from images by supplying flexible tools from image processing, computer vision
and graph theory bundled in a convenient package. Thus, NEFI constitutes a
scalable alternative to tedious and error-prone manual graph extraction and
special purpose tools. We anticipate NEFI to enable the collection of larger
datasets by reducing the time spent on graph extraction. The analysis of these
new datasets may open up the possibility to gain new insights into the
structure and function of various types of networks. NEFI is open source and
available http://nefi.mpi-inf.mpg.de.
| [
{
"version": "v1",
"created": "Wed, 18 Feb 2015 14:20:25 GMT"
}
] | 2015-02-19T00:00:00 | [
[
"Dirnberger",
"Michael",
""
],
[
"Neumann",
"Adrian",
""
],
[
"Kehl",
"Tim",
""
]
] | TITLE: NEFI: Network Extraction From Images
ABSTRACT: Networks and network-like structures are amongst the central building blocks
of many technological and biological systems. Given a mathematical graph
representation of a network, methods from graph theory enable a precise
investigation of its properties. Software for the analysis of graphs is widely
available and has been applied to graphs describing large scale networks such
as social networks, protein-interaction networks, etc. In these applications,
graph acquisition, i.e., the extraction of a mathematical graph from a network,
is relatively simple. However, for many network-like structures, e.g. leaf
venations, slime molds and mud cracks, data collection relies on images where
graph extraction requires domain-specific solutions or even manual. Here we
introduce Network Extraction From Images, NEFI, a software tool that
automatically extracts accurate graphs from images of a wide range of networks
originating in various domains. While there is previous work on graph
extraction from images, theoretical results are fully accessible only to an
expert audience and ready-to-use implementations for non-experts are rarely
available or insufficiently documented. NEFI provides a novel platform allowing
practitioners from many disciplines to easily extract graph representations
from images by supplying flexible tools from image processing, computer vision
and graph theory bundled in a convenient package. Thus, NEFI constitutes a
scalable alternative to tedious and error-prone manual graph extraction and
special purpose tools. We anticipate NEFI to enable the collection of larger
datasets by reducing the time spent on graph extraction. The analysis of these
new datasets may open up the possibility to gain new insights into the
structure and function of various types of networks. NEFI is open source and
available http://nefi.mpi-inf.mpg.de.
|
1502.04983 | Thanapong Intharah | Thanapong Intharah and Gabriel J. Brostow | Context Tricks for Cheap Semantic Segmentation | Supplementary material can be found at
http://www0.cs.ucl.ac.uk/staff/T.Intharah/research.html | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate semantic labeling of image pixels is difficult because intra-class
variability is often greater than inter-class variability. In turn, fast
semantic segmentation is hard because accurate models are usually too
complicated to also run quickly at test-time. Our experience with building and
running semantic segmentation systems has also shown a reasonably obvious
bottleneck on model complexity, imposed by small training datasets. We
therefore propose two simple complementary strategies that leverage context to
give better semantic segmentation, while scaling up or down to train on
different-sized datasets.
As easy modifications for existing semantic segmentation algorithms, we
introduce Decorrelated Semantic Texton Forests, and the Context Sensitive Image
Level Prior. The proposed modifications are tested using a Semantic Texton
Forest (STF) system, and the modifications are validated on two standard
benchmark datasets, MSRC-21 and PascalVOC-2010. In Python based comparisons,
our system is insignificantly slower than STF at test-time, yet produces
superior semantic segmentations overall, with just push-button training.
| [
{
"version": "v1",
"created": "Tue, 17 Feb 2015 18:08:53 GMT"
}
] | 2015-02-18T00:00:00 | [
[
"Intharah",
"Thanapong",
""
],
[
"Brostow",
"Gabriel J.",
""
]
] | TITLE: Context Tricks for Cheap Semantic Segmentation
ABSTRACT: Accurate semantic labeling of image pixels is difficult because intra-class
variability is often greater than inter-class variability. In turn, fast
semantic segmentation is hard because accurate models are usually too
complicated to also run quickly at test-time. Our experience with building and
running semantic segmentation systems has also shown a reasonably obvious
bottleneck on model complexity, imposed by small training datasets. We
therefore propose two simple complementary strategies that leverage context to
give better semantic segmentation, while scaling up or down to train on
different-sized datasets.
As easy modifications for existing semantic segmentation algorithms, we
introduce Decorrelated Semantic Texton Forests, and the Context Sensitive Image
Level Prior. The proposed modifications are tested using a Semantic Texton
Forest (STF) system, and the modifications are validated on two standard
benchmark datasets, MSRC-21 and PascalVOC-2010. In Python based comparisons,
our system is insignificantly slower than STF at test-time, yet produces
superior semantic segmentations overall, with just push-button training.
|
1212.0975 | Arya Iranmehr | Hamed Masnadi-Shirazi, Nuno Vasconcelos and Arya Iranmehr | Cost-Sensitive Support Vector Machines | 32 pages, 4 figures | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A new procedure for learning cost-sensitive SVM(CS-SVM) classifiers is
proposed. The SVM hinge loss is extended to the cost sensitive setting, and the
CS-SVM is derived as the minimizer of the associated risk. The extension of the
hinge loss draws on recent connections between risk minimization and
probability elicitation. These connections are generalized to cost-sensitive
classification, in a manner that guarantees consistency with the cost-sensitive
Bayes risk, and associated Bayes decision rule. This ensures that optimal
decision rules, under the new hinge loss, implement the Bayes-optimal
cost-sensitive classification boundary. Minimization of the new hinge loss is
shown to be a generalization of the classic SVM optimization problem, and can
be solved by identical procedures. The dual problem of CS-SVM is carefully
scrutinized by means of regularization theory and sensitivity analysis and the
CS-SVM algorithm is substantiated. The proposed algorithm is also extended to
cost-sensitive learning with example dependent costs. The minimum cost
sensitive risk is proposed as the performance measure and is connected to ROC
analysis through vector optimization. The resulting algorithm avoids the
shortcomings of previous approaches to cost-sensitive SVM design, and is shown
to have superior experimental performance on a large number of cost sensitive
and imbalanced datasets.
| [
{
"version": "v1",
"created": "Wed, 5 Dec 2012 09:24:11 GMT"
},
{
"version": "v2",
"created": "Sun, 15 Feb 2015 11:17:57 GMT"
}
] | 2015-02-17T00:00:00 | [
[
"Masnadi-Shirazi",
"Hamed",
""
],
[
"Vasconcelos",
"Nuno",
""
],
[
"Iranmehr",
"Arya",
""
]
] | TITLE: Cost-Sensitive Support Vector Machines
ABSTRACT: A new procedure for learning cost-sensitive SVM(CS-SVM) classifiers is
proposed. The SVM hinge loss is extended to the cost sensitive setting, and the
CS-SVM is derived as the minimizer of the associated risk. The extension of the
hinge loss draws on recent connections between risk minimization and
probability elicitation. These connections are generalized to cost-sensitive
classification, in a manner that guarantees consistency with the cost-sensitive
Bayes risk, and associated Bayes decision rule. This ensures that optimal
decision rules, under the new hinge loss, implement the Bayes-optimal
cost-sensitive classification boundary. Minimization of the new hinge loss is
shown to be a generalization of the classic SVM optimization problem, and can
be solved by identical procedures. The dual problem of CS-SVM is carefully
scrutinized by means of regularization theory and sensitivity analysis and the
CS-SVM algorithm is substantiated. The proposed algorithm is also extended to
cost-sensitive learning with example dependent costs. The minimum cost
sensitive risk is proposed as the performance measure and is connected to ROC
analysis through vector optimization. The resulting algorithm avoids the
shortcomings of previous approaches to cost-sensitive SVM design, and is shown
to have superior experimental performance on a large number of cost sensitive
and imbalanced datasets.
|
1309.2168 | Pablo Gonz\'alez-Brevis Dr | Jacek Gondzio, Pablo Gonz\'alez-Brevis and Pedro Munari | Large-scale optimization with the primal-dual column generation method | 28 pages, 1 figure, minor revision, scaled CPU times | null | null | Technical Report ERGO 13-014 | math.OC cs.LG cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The primal-dual column generation method (PDCGM) is a general-purpose column
generation technique that relies on the primal-dual interior point method to
solve the restricted master problems. The use of this interior point method
variant allows to obtain suboptimal and well-centered dual solutions which
naturally stabilizes the column generation. As recently presented in the
literature, reductions in the number of calls to the oracle and in the CPU
times are typically observed when compared to the standard column generation,
which relies on extreme optimal dual solutions. However, these results are
based on relatively small problems obtained from linear relaxations of
combinatorial applications. In this paper, we investigate the behaviour of the
PDCGM in a broader context, namely when solving large-scale convex optimization
problems. We have selected applications that arise in important real-life
contexts such as data analysis (multiple kernel learning problem),
decision-making under uncertainty (two-stage stochastic programming problems)
and telecommunication and transportation networks (multicommodity network flow
problem). In the numerical experiments, we use publicly available benchmark
instances to compare the performance of the PDCGM against recent results for
different methods presented in the literature, which were the best available
results to date. The analysis of these results suggests that the PDCGM offers
an attractive alternative over specialized methods since it remains competitive
in terms of number of iterations and CPU times even for large-scale
optimization problems.
| [
{
"version": "v1",
"created": "Mon, 9 Sep 2013 14:19:10 GMT"
},
{
"version": "v2",
"created": "Mon, 16 Feb 2015 17:40:28 GMT"
}
] | 2015-02-17T00:00:00 | [
[
"Gondzio",
"Jacek",
""
],
[
"González-Brevis",
"Pablo",
""
],
[
"Munari",
"Pedro",
""
]
] | TITLE: Large-scale optimization with the primal-dual column generation method
ABSTRACT: The primal-dual column generation method (PDCGM) is a general-purpose column
generation technique that relies on the primal-dual interior point method to
solve the restricted master problems. The use of this interior point method
variant allows to obtain suboptimal and well-centered dual solutions which
naturally stabilizes the column generation. As recently presented in the
literature, reductions in the number of calls to the oracle and in the CPU
times are typically observed when compared to the standard column generation,
which relies on extreme optimal dual solutions. However, these results are
based on relatively small problems obtained from linear relaxations of
combinatorial applications. In this paper, we investigate the behaviour of the
PDCGM in a broader context, namely when solving large-scale convex optimization
problems. We have selected applications that arise in important real-life
contexts such as data analysis (multiple kernel learning problem),
decision-making under uncertainty (two-stage stochastic programming problems)
and telecommunication and transportation networks (multicommodity network flow
problem). In the numerical experiments, we use publicly available benchmark
instances to compare the performance of the PDCGM against recent results for
different methods presented in the literature, which were the best available
results to date. The analysis of these results suggests that the PDCGM offers
an attractive alternative over specialized methods since it remains competitive
in terms of number of iterations and CPU times even for large-scale
optimization problems.
|
1406.7611 | Lutz Bornmann Dr. | Lutz Bornmann | Validity of altmetrics data for measuring societal impact: A study using
data from Altmetric and F1000Prime | null | null | null | null | cs.DL physics.soc-ph stat.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Can altmetric data be validly used for the measurement of societal impact?
The current study seeks to answer this question with a comprehensive dataset
(about 100,000 records) from very disparate sources (F1000, Altmetric, and an
in-house database based on Web of Science). In the F1000 peer review system,
experts attach particular tags to scientific papers which indicate whether a
paper could be of interest for science or rather for other segments of society.
The results show that papers with the tag "good for teaching" do achieve higher
altmetric counts than papers without this tag - if the quality of the papers is
controlled. At the same time, a higher citation count is shown especially by
papers with a tag that is specifically scientifically oriented ("new finding").
The findings indicate that papers tailored for a readership outside the area of
research should lead to societal impact. If altmetric data is to be used for
the measurement of societal impact, the question arises of its normalization.
In bibliometrics, citations are normalized for the papers' subject area and
publication year. This study has taken a second analytic step involving a
possible normalization of altmetric data. As the results show there are
particular scientific topics which are of especial interest for a wide
audience. Since these more or less interesting topics are not completely
reflected in Thomson Reuters' journal sets, a normalization of altmetric data
should not be based on the level of subject categories, but on the level of
topics.
| [
{
"version": "v1",
"created": "Mon, 30 Jun 2014 06:19:24 GMT"
},
{
"version": "v2",
"created": "Mon, 16 Feb 2015 08:37:42 GMT"
}
] | 2015-02-17T00:00:00 | [
[
"Bornmann",
"Lutz",
""
]
] | TITLE: Validity of altmetrics data for measuring societal impact: A study using
data from Altmetric and F1000Prime
ABSTRACT: Can altmetric data be validly used for the measurement of societal impact?
The current study seeks to answer this question with a comprehensive dataset
(about 100,000 records) from very disparate sources (F1000, Altmetric, and an
in-house database based on Web of Science). In the F1000 peer review system,
experts attach particular tags to scientific papers which indicate whether a
paper could be of interest for science or rather for other segments of society.
The results show that papers with the tag "good for teaching" do achieve higher
altmetric counts than papers without this tag - if the quality of the papers is
controlled. At the same time, a higher citation count is shown especially by
papers with a tag that is specifically scientifically oriented ("new finding").
The findings indicate that papers tailored for a readership outside the area of
research should lead to societal impact. If altmetric data is to be used for
the measurement of societal impact, the question arises of its normalization.
In bibliometrics, citations are normalized for the papers' subject area and
publication year. This study has taken a second analytic step involving a
possible normalization of altmetric data. As the results show there are
particular scientific topics which are of especial interest for a wide
audience. Since these more or less interesting topics are not completely
reflected in Thomson Reuters' journal sets, a normalization of altmetric data
should not be based on the level of subject categories, but on the level of
topics.
|
1409.2863 | Lutz Bornmann Dr. | Lutz Bornmann | Usefulness of altmetrics for measuring the broader impact of research: A
case study using data from PLOS (altmetrics) and F1000Prime (paper tags) | arXiv admin note: text overlap with arXiv:1406.7611 | null | null | null | cs.DL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Purpose: Whereas citation counts allow the measurement of the impact of
research on research itself, an important role in the measurement of the impact
of research on other parts of society is ascribed to altmetrics. The present
case study investigates the usefulness of altmetrics for measuring the broader
impact of research. Methods: This case study is essentially based on a dataset
with papers obtained from F1000. The dataset was augmented with altmetrics
(such as Twitter counts) which were provided by PLOS (the Public Library of
Science). In total, the case study covers a total of 1,082 papers. Findings:
The F1000 dataset contains tags on papers which were assigned intellectually by
experts and which can characterise a paper. The most interesting tag for
altmetric research is "good for teaching". This tag is assigned to papers which
could be of interest to a wider circle of readers than the peers in a
specialist area. Particularly on Facebook and Twitter, one could expect papers
with this tag to be mentioned more often than those without this tag. With
respect to the "good for teaching" tag, the results from regression models were
able to confirm these expectations: Papers with this tag show significantly
higher Facebook and Twitter counts than papers without this tag. This
association could not be seen with Mendeley or Figshare counts (that is with
counts from platforms which are chiefly of interest in a scientific context).
Conclusions: The results of the current study indicate that Facebook and
Twitter, but not Figshare or Mendeley, can provide indications of papers which
are of interest to a broader circle of readers (and not only for the peers in a
specialist area), and seem therefore be useful for societal impact measurement.
| [
{
"version": "v1",
"created": "Tue, 9 Sep 2014 08:05:37 GMT"
},
{
"version": "v2",
"created": "Mon, 16 Feb 2015 08:43:56 GMT"
}
] | 2015-02-17T00:00:00 | [
[
"Bornmann",
"Lutz",
""
]
] | TITLE: Usefulness of altmetrics for measuring the broader impact of research: A
case study using data from PLOS (altmetrics) and F1000Prime (paper tags)
ABSTRACT: Purpose: Whereas citation counts allow the measurement of the impact of
research on research itself, an important role in the measurement of the impact
of research on other parts of society is ascribed to altmetrics. The present
case study investigates the usefulness of altmetrics for measuring the broader
impact of research. Methods: This case study is essentially based on a dataset
with papers obtained from F1000. The dataset was augmented with altmetrics
(such as Twitter counts) which were provided by PLOS (the Public Library of
Science). In total, the case study covers a total of 1,082 papers. Findings:
The F1000 dataset contains tags on papers which were assigned intellectually by
experts and which can characterise a paper. The most interesting tag for
altmetric research is "good for teaching". This tag is assigned to papers which
could be of interest to a wider circle of readers than the peers in a
specialist area. Particularly on Facebook and Twitter, one could expect papers
with this tag to be mentioned more often than those without this tag. With
respect to the "good for teaching" tag, the results from regression models were
able to confirm these expectations: Papers with this tag show significantly
higher Facebook and Twitter counts than papers without this tag. This
association could not be seen with Mendeley or Figshare counts (that is with
counts from platforms which are chiefly of interest in a scientific context).
Conclusions: The results of the current study indicate that Facebook and
Twitter, but not Figshare or Mendeley, can provide indications of papers which
are of interest to a broader circle of readers (and not only for the peers in a
specialist area), and seem therefore be useful for societal impact measurement.
|
1502.03536 | Vamsi Ithapu | Chris Hinrichs, Vamsi K Ithapu, Qinyuan Sun, Sterling C Johnson, Vikas
Singh | Speeding up Permutation Testing in Neuroimaging | NIPS 13 | Advances in neural information processing systems (2013), pp.
890-898 | null | null | stat.CO cs.AI stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multiple hypothesis testing is a significant problem in nearly all
neuroimaging studies. In order to correct for this phenomena, we require a
reliable estimate of the Family-Wise Error Rate (FWER). The well known
Bonferroni correction method, while simple to implement, is quite conservative,
and can substantially under-power a study because it ignores dependencies
between test statistics. Permutation testing, on the other hand, is an exact,
non-parametric method of estimating the FWER for a given $\alpha$-threshold,
but for acceptably low thresholds the computational burden can be prohibitive.
In this paper, we show that permutation testing in fact amounts to populating
the columns of a very large matrix ${\bf P}$. By analyzing the spectrum of this
matrix, under certain conditions, we see that ${\bf P}$ has a low-rank plus a
low-variance residual decomposition which makes it suitable for highly
sub--sampled --- on the order of $0.5\%$ --- matrix completion methods. Based
on this observation, we propose a novel permutation testing methodology which
offers a large speedup, without sacrificing the fidelity of the estimated FWER.
Our evaluations on four different neuroimaging datasets show that a
computational speedup factor of roughly $50\times$ can be achieved while
recovering the FWER distribution up to very high accuracy. Further, we show
that the estimated $\alpha$-threshold is also recovered faithfully, and is
stable.
| [
{
"version": "v1",
"created": "Thu, 12 Feb 2015 04:30:06 GMT"
}
] | 2015-02-17T00:00:00 | [
[
"Hinrichs",
"Chris",
""
],
[
"Ithapu",
"Vamsi K",
""
],
[
"Sun",
"Qinyuan",
""
],
[
"Johnson",
"Sterling C",
""
],
[
"Singh",
"Vikas",
""
]
] | TITLE: Speeding up Permutation Testing in Neuroimaging
ABSTRACT: Multiple hypothesis testing is a significant problem in nearly all
neuroimaging studies. In order to correct for this phenomena, we require a
reliable estimate of the Family-Wise Error Rate (FWER). The well known
Bonferroni correction method, while simple to implement, is quite conservative,
and can substantially under-power a study because it ignores dependencies
between test statistics. Permutation testing, on the other hand, is an exact,
non-parametric method of estimating the FWER for a given $\alpha$-threshold,
but for acceptably low thresholds the computational burden can be prohibitive.
In this paper, we show that permutation testing in fact amounts to populating
the columns of a very large matrix ${\bf P}$. By analyzing the spectrum of this
matrix, under certain conditions, we see that ${\bf P}$ has a low-rank plus a
low-variance residual decomposition which makes it suitable for highly
sub--sampled --- on the order of $0.5\%$ --- matrix completion methods. Based
on this observation, we propose a novel permutation testing methodology which
offers a large speedup, without sacrificing the fidelity of the estimated FWER.
Our evaluations on four different neuroimaging datasets show that a
computational speedup factor of roughly $50\times$ can be achieved while
recovering the FWER distribution up to very high accuracy. Further, we show
that the estimated $\alpha$-threshold is also recovered faithfully, and is
stable.
|
1502.04132 | Zhenzhong Lan | Zhenzhong Lan, Xuanchong Li, Ming Lin, Alexander G. Hauptmann | Long-short Term Motion Feature for Action Classification and Retrieval | arXiv admin note: text overlap with arXiv:1411.6660 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a method for representing motion information for video
classification and retrieval. We improve upon local descriptor based methods
that have been among the most popular and successful models for representing
videos. The desired local descriptors need to satisfy two requirements: 1) to
be representative, 2) to be discriminative. Therefore, they need to occur
frequently enough in the videos and to be be able to tell the difference among
different types of motions. To generate such local descriptors, the video
blocks they are based on must contain just the right amount of motion
information. However, current state-of-the-art local descriptor methods use
video blocks with a single fixed size, which is insufficient for covering
actions with varying speeds. In this paper, we introduce a long-short term
motion feature that generates descriptors from video blocks with multiple
lengths, thus covering motions with large speed variance. Experimental results
show that, albeit simple, our model achieves state-of-the-arts results on
several benchmark datasets.
| [
{
"version": "v1",
"created": "Fri, 13 Feb 2015 21:15:57 GMT"
}
] | 2015-02-17T00:00:00 | [
[
"Lan",
"Zhenzhong",
""
],
[
"Li",
"Xuanchong",
""
],
[
"Lin",
"Ming",
""
],
[
"Hauptmann",
"Alexander G.",
""
]
] | TITLE: Long-short Term Motion Feature for Action Classification and Retrieval
ABSTRACT: We propose a method for representing motion information for video
classification and retrieval. We improve upon local descriptor based methods
that have been among the most popular and successful models for representing
videos. The desired local descriptors need to satisfy two requirements: 1) to
be representative, 2) to be discriminative. Therefore, they need to occur
frequently enough in the videos and to be be able to tell the difference among
different types of motions. To generate such local descriptors, the video
blocks they are based on must contain just the right amount of motion
information. However, current state-of-the-art local descriptor methods use
video blocks with a single fixed size, which is insufficient for covering
actions with varying speeds. In this paper, we introduce a long-short term
motion feature that generates descriptors from video blocks with multiple
lengths, thus covering motions with large speed variance. Experimental results
show that, albeit simple, our model achieves state-of-the-arts results on
several benchmark datasets.
|
1502.04220 | Can Lu | Can Lu, Jeffrey Xu Yu, Rong-Hua Li, Hao Wei | Exploring Hierarchies in Online Social Networks | null | null | null | null | cs.SI cs.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Social hierarchy (i.e., pyramid structure of societies) is a fundamental
concept in sociology and social network analysis. The importance of social
hierarchy in a social network is that the topological structure of the social
hierarchy is essential in both shaping the nature of social interactions
between individuals and unfolding the structure of the social networks. The
social hierarchy found in a social network can be utilized to improve the
accuracy of link prediction, provide better query results, rank web pages, and
study information flow and spread in complex networks. In this paper, we model
a social network as a directed graph G, and consider the social hierarchy as
DAG (directed acyclic graph) of G, denoted as GD. By DAG, all the vertices in G
can be partitioned into different levels, the vertices at the same level
represent a disjoint group in the social hierarchy, and all the edges in DAG
follow one direction. The main issue we study in this paper is how to find DAG
GD in G. The approach we take is to find GD by removing all possible cycles
from G such that G = U(G) + GD where U(G) is a maximum Eulerian subgraph which
contains all possible cycles. We give the reasons for doing so, investigate the
properties of GD found, and discuss the applications. In addition, we develop a
novel two-phase algorithm, called Greedy-&-Refine, which greedily computes an
Eulerian subgraph and then refines this greedy solution to find the maximum
Eulerian subgraph. We give a bound between the greedy solution and the optimal.
The quality of our greedy approach is high. We conduct comprehensive
experimental studies over 14 real-world datasets. The results show that our
algorithms are at least two orders of magnitude faster than the baseline
algorithm.
| [
{
"version": "v1",
"created": "Sat, 14 Feb 2015 16:02:00 GMT"
}
] | 2015-02-17T00:00:00 | [
[
"Lu",
"Can",
""
],
[
"Yu",
"Jeffrey Xu",
""
],
[
"Li",
"Rong-Hua",
""
],
[
"Wei",
"Hao",
""
]
] | TITLE: Exploring Hierarchies in Online Social Networks
ABSTRACT: Social hierarchy (i.e., pyramid structure of societies) is a fundamental
concept in sociology and social network analysis. The importance of social
hierarchy in a social network is that the topological structure of the social
hierarchy is essential in both shaping the nature of social interactions
between individuals and unfolding the structure of the social networks. The
social hierarchy found in a social network can be utilized to improve the
accuracy of link prediction, provide better query results, rank web pages, and
study information flow and spread in complex networks. In this paper, we model
a social network as a directed graph G, and consider the social hierarchy as
DAG (directed acyclic graph) of G, denoted as GD. By DAG, all the vertices in G
can be partitioned into different levels, the vertices at the same level
represent a disjoint group in the social hierarchy, and all the edges in DAG
follow one direction. The main issue we study in this paper is how to find DAG
GD in G. The approach we take is to find GD by removing all possible cycles
from G such that G = U(G) + GD where U(G) is a maximum Eulerian subgraph which
contains all possible cycles. We give the reasons for doing so, investigate the
properties of GD found, and discuss the applications. In addition, we develop a
novel two-phase algorithm, called Greedy-&-Refine, which greedily computes an
Eulerian subgraph and then refines this greedy solution to find the maximum
Eulerian subgraph. We give a bound between the greedy solution and the optimal.
The quality of our greedy approach is high. We conduct comprehensive
experimental studies over 14 real-world datasets. The results show that our
algorithms are at least two orders of magnitude faster than the baseline
algorithm.
|
1310.6670 | Matteo Camilli M.Sc. | Carlo Bellettini, Matteo Camilli, Lorenzo Capra, Mattia Monga | Distributed CTL Model Checking in the Cloud | 8 pages | null | 10.1109/SYNASC.2014.52 | null | cs.SE cs.DC cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The recent extensive availability of "big data" platforms calls for a more
widespread adoption by the formal verification community. In fact, formal
verification requires high performance data processing software for extracting
knowledge from the unprecedented amount of data which come from analyzed
systems. Since cloud based computing resources have became easily accessible,
there is an opportunity for verification techniques and tools to undergo a deep
technological transition to exploit the new available architectures. This has
created an increasing interest in parallelizing and distributing verification
techniques. In this paper we introduce a distributed approach which exploits
techniques typically used by the "big data" community to enable verification of
Computation Tree Logic (CTL) formulas on very large state spaces using
distributed systems and cloud computing facilities. The outcome of several
tests performed on benchmark specifications are presented, thus showing the
convenience of the proposed approach.
| [
{
"version": "v1",
"created": "Thu, 24 Oct 2013 16:54:37 GMT"
}
] | 2015-02-16T00:00:00 | [
[
"Bellettini",
"Carlo",
""
],
[
"Camilli",
"Matteo",
""
],
[
"Capra",
"Lorenzo",
""
],
[
"Monga",
"Mattia",
""
]
] | TITLE: Distributed CTL Model Checking in the Cloud
ABSTRACT: The recent extensive availability of "big data" platforms calls for a more
widespread adoption by the formal verification community. In fact, formal
verification requires high performance data processing software for extracting
knowledge from the unprecedented amount of data which come from analyzed
systems. Since cloud based computing resources have became easily accessible,
there is an opportunity for verification techniques and tools to undergo a deep
technological transition to exploit the new available architectures. This has
created an increasing interest in parallelizing and distributing verification
techniques. In this paper we introduce a distributed approach which exploits
techniques typically used by the "big data" community to enable verification of
Computation Tree Logic (CTL) formulas on very large state spaces using
distributed systems and cloud computing facilities. The outcome of several
tests performed on benchmark specifications are presented, thus showing the
convenience of the proposed approach.
|
1408.3873 | Lorenzo Livi | Lorenzo Livi, Antonello Rizzi, Alireza Sadeghian | Classifying sequences by the optimized dissimilarity space embedding
approach: a case study on the solubility analysis of the E. coli proteome | 10 pages, 49 references | null | 10.3233/IFS-151550 | null | cs.CV cs.AI physics.bio-ph q-bio.BM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We evaluate a version of the recently-proposed classification system named
Optimized Dissimilarity Space Embedding (ODSE) that operates in the input space
of sequences of generic objects. The ODSE system has been originally presented
as a classification system for patterns represented as labeled graphs. However,
since ODSE is founded on the dissimilarity space representation of the input
data, the classifier can be easily adapted to any input domain where it is
possible to define a meaningful dissimilarity measure. Here we demonstrate the
effectiveness of the ODSE classifier for sequences by considering an
application dealing with the recognition of the solubility degree of the
Escherichia coli proteome. Solubility, or analogously aggregation propensity,
is an important property of protein molecules, which is intimately related to
the mechanisms underlying the chemico-physical process of folding. Each protein
of our dataset is initially associated with a solubility degree and it is
represented as a sequence of symbols, denoting the 20 amino acid residues. The
herein obtained computational results, which we stress that have been achieved
with no context-dependent tuning of the ODSE system, confirm the validity and
generality of the ODSE-based approach for structured data classification.
| [
{
"version": "v1",
"created": "Sun, 17 Aug 2014 23:46:55 GMT"
},
{
"version": "v2",
"created": "Wed, 14 Jan 2015 21:20:19 GMT"
}
] | 2015-02-16T00:00:00 | [
[
"Livi",
"Lorenzo",
""
],
[
"Rizzi",
"Antonello",
""
],
[
"Sadeghian",
"Alireza",
""
]
] | TITLE: Classifying sequences by the optimized dissimilarity space embedding
approach: a case study on the solubility analysis of the E. coli proteome
ABSTRACT: We evaluate a version of the recently-proposed classification system named
Optimized Dissimilarity Space Embedding (ODSE) that operates in the input space
of sequences of generic objects. The ODSE system has been originally presented
as a classification system for patterns represented as labeled graphs. However,
since ODSE is founded on the dissimilarity space representation of the input
data, the classifier can be easily adapted to any input domain where it is
possible to define a meaningful dissimilarity measure. Here we demonstrate the
effectiveness of the ODSE classifier for sequences by considering an
application dealing with the recognition of the solubility degree of the
Escherichia coli proteome. Solubility, or analogously aggregation propensity,
is an important property of protein molecules, which is intimately related to
the mechanisms underlying the chemico-physical process of folding. Each protein
of our dataset is initially associated with a solubility degree and it is
represented as a sequence of symbols, denoting the 20 amino acid residues. The
herein obtained computational results, which we stress that have been achieved
with no context-dependent tuning of the ODSE system, confirm the validity and
generality of the ODSE-based approach for structured data classification.
|
1409.2802 | William March | William B. March and George Biros | Far-Field Compression for Fast Kernel Summation Methods in High
Dimensions | 43 pages, 21 figures | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider fast kernel summations in high dimensions: given a large set of
points in $d$ dimensions (with $d \gg 3$) and a pair-potential function (the
{\em kernel} function), we compute a weighted sum of all pairwise kernel
interactions for each point in the set. Direct summation is equivalent to a
(dense) matrix-vector multiplication and scales quadratically with the number
of points. Fast kernel summation algorithms reduce this cost to log-linear or
linear complexity.
Treecodes and Fast Multipole Methods (FMMs) deliver tremendous speedups by
constructing approximate representations of interactions of points that are far
from each other. In algebraic terms, these representations correspond to
low-rank approximations of blocks of the overall interaction matrix. Existing
approaches require an excessive number of kernel evaluations with increasing
$d$ and number of points in the dataset.
To address this issue, we use a randomized algebraic approach in which we
first sample the rows of a block and then construct its approximate, low-rank
interpolative decomposition. We examine the feasibility of this approach
theoretically and experimentally. We provide a new theoretical result showing a
tighter bound on the reconstruction error from uniformly sampling rows than the
existing state-of-the-art. We demonstrate that our sampling approach is
competitive with existing (but prohibitively expensive) methods from the
literature. We also construct kernel matrices for the Laplacian, Gaussian, and
polynomial kernels -- all commonly used in physics and data analysis. We
explore the numerical properties of blocks of these matrices, and show that
they are amenable to our approach. Depending on the data set, our randomized
algorithm can successfully compute low rank approximations in high dimensions.
We report results for data sets with ambient dimensions from four to 1,000.
| [
{
"version": "v1",
"created": "Tue, 9 Sep 2014 16:28:40 GMT"
},
{
"version": "v2",
"created": "Fri, 13 Feb 2015 00:30:28 GMT"
}
] | 2015-02-16T00:00:00 | [
[
"March",
"William B.",
""
],
[
"Biros",
"George",
""
]
] | TITLE: Far-Field Compression for Fast Kernel Summation Methods in High
Dimensions
ABSTRACT: We consider fast kernel summations in high dimensions: given a large set of
points in $d$ dimensions (with $d \gg 3$) and a pair-potential function (the
{\em kernel} function), we compute a weighted sum of all pairwise kernel
interactions for each point in the set. Direct summation is equivalent to a
(dense) matrix-vector multiplication and scales quadratically with the number
of points. Fast kernel summation algorithms reduce this cost to log-linear or
linear complexity.
Treecodes and Fast Multipole Methods (FMMs) deliver tremendous speedups by
constructing approximate representations of interactions of points that are far
from each other. In algebraic terms, these representations correspond to
low-rank approximations of blocks of the overall interaction matrix. Existing
approaches require an excessive number of kernel evaluations with increasing
$d$ and number of points in the dataset.
To address this issue, we use a randomized algebraic approach in which we
first sample the rows of a block and then construct its approximate, low-rank
interpolative decomposition. We examine the feasibility of this approach
theoretically and experimentally. We provide a new theoretical result showing a
tighter bound on the reconstruction error from uniformly sampling rows than the
existing state-of-the-art. We demonstrate that our sampling approach is
competitive with existing (but prohibitively expensive) methods from the
literature. We also construct kernel matrices for the Laplacian, Gaussian, and
polynomial kernels -- all commonly used in physics and data analysis. We
explore the numerical properties of blocks of these matrices, and show that
they are amenable to our approach. Depending on the data set, our randomized
algorithm can successfully compute low rank approximations in high dimensions.
We report results for data sets with ambient dimensions from four to 1,000.
|
1502.03845 | Alessandro Provetti | Biagio Bonasera, Emilio Ferrara, Giacomo Fiumara, Francesco Pagano,
Alessandro Provetti | Adaptive Search over Sorted Sets | 9 pages | Journal of Discrete Algorithms, Volume 30, 2015, pp. 128--133 | 10.1016/j.jda.2014.12.007 | null | cs.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We revisit the classical algorithms for searching over sorted sets to
introduce an algorithm refinement, called Adaptive Search, that combines the
good features of Interpolation search and those of Binary search. W.r.t.
Interpolation search, only a constant number of extra comparisons is
introduced. Yet, under diverse input data distributions our algorithm shows
costs comparable to that of Interpolation search, i.e., O(log log n) while the
worst-case cost is always in O(log n), as with Binary search. On benchmarks
drawn from large datasets, both synthetic and real-life, Adaptive search scores
better times and lesser memory accesses even than Santoro and Sidney's
Interpolation-Binary search.
| [
{
"version": "v1",
"created": "Thu, 12 Feb 2015 22:12:54 GMT"
}
] | 2015-02-16T00:00:00 | [
[
"Bonasera",
"Biagio",
""
],
[
"Ferrara",
"Emilio",
""
],
[
"Fiumara",
"Giacomo",
""
],
[
"Pagano",
"Francesco",
""
],
[
"Provetti",
"Alessandro",
""
]
] | TITLE: Adaptive Search over Sorted Sets
ABSTRACT: We revisit the classical algorithms for searching over sorted sets to
introduce an algorithm refinement, called Adaptive Search, that combines the
good features of Interpolation search and those of Binary search. W.r.t.
Interpolation search, only a constant number of extra comparisons is
introduced. Yet, under diverse input data distributions our algorithm shows
costs comparable to that of Interpolation search, i.e., O(log log n) while the
worst-case cost is always in O(log n), as with Binary search. On benchmarks
drawn from large datasets, both synthetic and real-life, Adaptive search scores
better times and lesser memory accesses even than Santoro and Sidney's
Interpolation-Binary search.
|
1502.03851 | Arash Vahdat | Mehran Khodabandeh, Arash Vahdat, Guang-Tong Zhou, Hossein
Hajimirsadeghi, Mehrsan Javan Roshtkhari, Greg Mori, Stephen Se | Discovering Human Interactions in Videos with Limited Data Labeling | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a novel approach for discovering human interactions in videos.
Activity understanding techniques usually require a large number of labeled
examples, which are not available in many practical cases. Here, we focus on
recovering semantically meaningful clusters of human-human and human-object
interaction in an unsupervised fashion. A new iterative solution is introduced
based on Maximum Margin Clustering (MMC), which also accepts user feedback to
refine clusters. This is achieved by formulating the whole process as a unified
constrained latent max-margin clustering problem. Extensive experiments have
been carried out over three challenging datasets, Collective Activity, VIRAT,
and UT-interaction. Empirical results demonstrate that the proposed algorithm
can efficiently discover perfect semantic clusters of human interactions with
only a small amount of labeling effort.
| [
{
"version": "v1",
"created": "Thu, 12 Feb 2015 22:38:28 GMT"
}
] | 2015-02-16T00:00:00 | [
[
"Khodabandeh",
"Mehran",
""
],
[
"Vahdat",
"Arash",
""
],
[
"Zhou",
"Guang-Tong",
""
],
[
"Hajimirsadeghi",
"Hossein",
""
],
[
"Roshtkhari",
"Mehrsan Javan",
""
],
[
"Mori",
"Greg",
""
],
[
"Se",
"Stephen",
""
]
] | TITLE: Discovering Human Interactions in Videos with Limited Data Labeling
ABSTRACT: We present a novel approach for discovering human interactions in videos.
Activity understanding techniques usually require a large number of labeled
examples, which are not available in many practical cases. Here, we focus on
recovering semantically meaningful clusters of human-human and human-object
interaction in an unsupervised fashion. A new iterative solution is introduced
based on Maximum Margin Clustering (MMC), which also accepts user feedback to
refine clusters. This is achieved by formulating the whole process as a unified
constrained latent max-margin clustering problem. Extensive experiments have
been carried out over three challenging datasets, Collective Activity, VIRAT,
and UT-interaction. Empirical results demonstrate that the proposed algorithm
can efficiently discover perfect semantic clusters of human interactions with
only a small amount of labeling effort.
|
1502.03879 | Weiya Ren | Weiya Ren | Semi-supervised Data Representation via Affinity Graph Learning | 10 pages,2 Tables. Written in Aug,2013 | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the general problem of utilizing both labeled and unlabeled data
to improve data representation performance. A new semi-supervised learning
framework is proposed by combing manifold regularization and data
representation methods such as Non negative matrix factorization and sparse
coding. We adopt unsupervised data representation methods as the learning
machines because they do not depend on the labeled data, which can improve
machine's generation ability as much as possible. The proposed framework forms
the Laplacian regularizer through learning the affinity graph. We incorporate
the new Laplacian regularizer into the unsupervised data representation to
smooth the low dimensional representation of data and make use of label
information. Experimental results on several real benchmark datasets indicate
that our semi-supervised learning framework achieves encouraging results
compared with state-of-art methods.
| [
{
"version": "v1",
"created": "Fri, 13 Feb 2015 03:35:15 GMT"
}
] | 2015-02-16T00:00:00 | [
[
"Ren",
"Weiya",
""
]
] | TITLE: Semi-supervised Data Representation via Affinity Graph Learning
ABSTRACT: We consider the general problem of utilizing both labeled and unlabeled data
to improve data representation performance. A new semi-supervised learning
framework is proposed by combing manifold regularization and data
representation methods such as Non negative matrix factorization and sparse
coding. We adopt unsupervised data representation methods as the learning
machines because they do not depend on the labeled data, which can improve
machine's generation ability as much as possible. The proposed framework forms
the Laplacian regularizer through learning the affinity graph. We incorporate
the new Laplacian regularizer into the unsupervised data representation to
smooth the low dimensional representation of data and make use of label
information. Experimental results on several real benchmark datasets indicate
that our semi-supervised learning framework achieves encouraging results
compared with state-of-art methods.
|
1502.04025 | Tilo Wettig | Paul Arts, Jacques Bloch, Peter Georg, Benjamin Glaessle, Simon
Heybrock, Yu Komatsubara, Robert Lohmayer, Simon Mages, Bernhard Mendl, Nils
Meyer, Alessio Parcianello, Dirk Pleiter, Florian Rappl, Mauro Rossi, Stefan
Solbrig, Giampietro Tecchiolli, Tilo Wettig, Gianpaolo Zanier | QPACE 2 and Domain Decomposition on the Intel Xeon Phi | plenary talk at Lattice 2014, to appear in the conference proceedings
PoS(LATTICE2014), 15 pages, 9 figures | null | null | null | cs.DC hep-lat physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We give an overview of QPACE 2, which is a custom-designed supercomputer
based on Intel Xeon Phi processors, developed in a collaboration of Regensburg
University and Eurotech. We give some general recommendations for how to write
high-performance code for the Xeon Phi and then discuss our implementation of a
domain-decomposition-based solver and present a number of benchmarks.
| [
{
"version": "v1",
"created": "Fri, 13 Feb 2015 15:23:52 GMT"
}
] | 2015-02-16T00:00:00 | [
[
"Arts",
"Paul",
""
],
[
"Bloch",
"Jacques",
""
],
[
"Georg",
"Peter",
""
],
[
"Glaessle",
"Benjamin",
""
],
[
"Heybrock",
"Simon",
""
],
[
"Komatsubara",
"Yu",
""
],
[
"Lohmayer",
"Robert",
""
],
[
"Mages",
"Simon",
""
],
[
"Mendl",
"Bernhard",
""
],
[
"Meyer",
"Nils",
""
],
[
"Parcianello",
"Alessio",
""
],
[
"Pleiter",
"Dirk",
""
],
[
"Rappl",
"Florian",
""
],
[
"Rossi",
"Mauro",
""
],
[
"Solbrig",
"Stefan",
""
],
[
"Tecchiolli",
"Giampietro",
""
],
[
"Wettig",
"Tilo",
""
],
[
"Zanier",
"Gianpaolo",
""
]
] | TITLE: QPACE 2 and Domain Decomposition on the Intel Xeon Phi
ABSTRACT: We give an overview of QPACE 2, which is a custom-designed supercomputer
based on Intel Xeon Phi processors, developed in a collaboration of Regensburg
University and Eurotech. We give some general recommendations for how to write
high-performance code for the Xeon Phi and then discuss our implementation of a
domain-decomposition-based solver and present a number of benchmarks.
|
1402.3144 | Marc Claesen | Marc Claesen, Frank De Smet, Johan A. K. Suykens, Bart De Moor | A Robust Ensemble Approach to Learn From Positive and Unlabeled Data
Using SVM Base Models | 34 pages, 6 figures, 6 tables. Accepted for publication in
Neurocomputing: Special Issue on Advances in Learning with Label Noise | null | 10.1016/j.neucom.2014.10.081 | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a novel approach to learn binary classifiers when only positive
and unlabeled instances are available (PU learning). This problem is routinely
cast as a supervised task with label noise in the negative set. We use an
ensemble of SVM models trained on bootstrap resamples of the training data for
increased robustness against label noise. The approach can be considered in a
bagging framework which provides an intuitive explanation for its mechanics in
a semi-supervised setting. We compared our method to state-of-the-art
approaches in simulations using multiple public benchmark data sets. The
included benchmark comprises three settings with increasing label noise: (i)
fully supervised, (ii) PU learning and (iii) PU learning with false positives.
Our approach shows a marginal improvement over existing methods in the second
setting and a significant improvement in the third.
| [
{
"version": "v1",
"created": "Thu, 13 Feb 2014 14:18:17 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Oct 2014 12:29:58 GMT"
}
] | 2015-02-13T00:00:00 | [
[
"Claesen",
"Marc",
""
],
[
"De Smet",
"Frank",
""
],
[
"Suykens",
"Johan A. K.",
""
],
[
"De Moor",
"Bart",
""
]
] | TITLE: A Robust Ensemble Approach to Learn From Positive and Unlabeled Data
Using SVM Base Models
ABSTRACT: We present a novel approach to learn binary classifiers when only positive
and unlabeled instances are available (PU learning). This problem is routinely
cast as a supervised task with label noise in the negative set. We use an
ensemble of SVM models trained on bootstrap resamples of the training data for
increased robustness against label noise. The approach can be considered in a
bagging framework which provides an intuitive explanation for its mechanics in
a semi-supervised setting. We compared our method to state-of-the-art
approaches in simulations using multiple public benchmark data sets. The
included benchmark comprises three settings with increasing label noise: (i)
fully supervised, (ii) PU learning and (iii) PU learning with false positives.
Our approach shows a marginal improvement over existing methods in the second
setting and a significant improvement in the third.
|
1408.6615 | Shervin Minaee | Shervin Minaee and AmirAli Abdolrashidi | Multispectral Palmprint Recognition Using Textural Features | 5 pages, Published in IEEE Signal Processing in Medicine and Biology
Symposium 2014 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In order to utilize identification to the best extent, we need robust and
fast algorithms and systems to process the data. Having palmprint as a reliable
and unique characteristic of every person, we extract and use its features
based on its geometry, lines and angles. There are countless ways to define
measures for the recognition task. To analyze a new point of view, we extracted
textural features and used them for palmprint recognition. Co-occurrence matrix
can be used for textural feature extraction. As classifiers, we have used the
minimum distance classifier (MDC) and the weighted majority voting system
(WMV). The proposed method is tested on a well-known multispectral palmprint
dataset of 6000 samples and an accuracy rate of 99.96-100% is obtained for most
scenarios which outperforms all previous works in multispectral palmprint
recognition.
| [
{
"version": "v1",
"created": "Thu, 28 Aug 2014 03:20:38 GMT"
},
{
"version": "v2",
"created": "Mon, 1 Sep 2014 04:49:30 GMT"
},
{
"version": "v3",
"created": "Thu, 12 Feb 2015 03:03:02 GMT"
}
] | 2015-02-13T00:00:00 | [
[
"Minaee",
"Shervin",
""
],
[
"Abdolrashidi",
"AmirAli",
""
]
] | TITLE: Multispectral Palmprint Recognition Using Textural Features
ABSTRACT: In order to utilize identification to the best extent, we need robust and
fast algorithms and systems to process the data. Having palmprint as a reliable
and unique characteristic of every person, we extract and use its features
based on its geometry, lines and angles. There are countless ways to define
measures for the recognition task. To analyze a new point of view, we extracted
textural features and used them for palmprint recognition. Co-occurrence matrix
can be used for textural feature extraction. As classifiers, we have used the
minimum distance classifier (MDC) and the weighted majority voting system
(WMV). The proposed method is tested on a well-known multispectral palmprint
dataset of 6000 samples and an accuracy rate of 99.96-100% is obtained for most
scenarios which outperforms all previous works in multispectral palmprint
recognition.
|
1502.03556 | Md. Hanif Seddiqui | Md. Hanif Seddiqui, Rudra Pratap Deb Nath, Masaki Aono | An Efficient Metric of Automatic Weight Generation for Properties in
Instance Matching Technique | 17 pages, 5 figures, 3 tables, pp. 1-17, publication year 2015,
journal publication, vol. 6 number 1 | Journal of Web and Semantic Technology (IJWeST), vol.6 no.1, pp.
1-17 (2015) | 10.5121/ijwest.2015.6101 | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The proliferation of heterogeneous data sources of semantic knowledge base
intensifies the need of an automatic instance matching technique. However, the
efficiency of instance matching is often influenced by the weight of a property
associated to instances. Automatic weight generation is a non-trivial, however
an important task in instance matching technique. Therefore, identifying an
appropriate metric for generating weight for a property automatically is
nevertheless a formidable task. In this paper, we investigate an approach of
generating weights automatically by considering hypotheses: (1) the weight of a
property is directly proportional to the ratio of the number of its distinct
values to the number of instances contain the property, and (2) the weight is
also proportional to the ratio of the number of distinct values of a property
to the number of instances in a training dataset. The basic intuition behind
the use of our approach is the classical theory of information content that
infrequent words are more informative than frequent ones. Our mathematical
model derives a metric for generating property weights automatically, which is
applied in instance matching system to produce re-conciliated instances
efficiently. Our experiments and evaluations show the effectiveness of our
proposed metric of automatic weight generation for properties in an instance
matching technique.
| [
{
"version": "v1",
"created": "Thu, 12 Feb 2015 07:51:39 GMT"
}
] | 2015-02-13T00:00:00 | [
[
"Seddiqui",
"Md. Hanif",
""
],
[
"Nath",
"Rudra Pratap Deb",
""
],
[
"Aono",
"Masaki",
""
]
] | TITLE: An Efficient Metric of Automatic Weight Generation for Properties in
Instance Matching Technique
ABSTRACT: The proliferation of heterogeneous data sources of semantic knowledge base
intensifies the need of an automatic instance matching technique. However, the
efficiency of instance matching is often influenced by the weight of a property
associated to instances. Automatic weight generation is a non-trivial, however
an important task in instance matching technique. Therefore, identifying an
appropriate metric for generating weight for a property automatically is
nevertheless a formidable task. In this paper, we investigate an approach of
generating weights automatically by considering hypotheses: (1) the weight of a
property is directly proportional to the ratio of the number of its distinct
values to the number of instances contain the property, and (2) the weight is
also proportional to the ratio of the number of distinct values of a property
to the number of instances in a training dataset. The basic intuition behind
the use of our approach is the classical theory of information content that
infrequent words are more informative than frequent ones. Our mathematical
model derives a metric for generating property weights automatically, which is
applied in instance matching system to produce re-conciliated instances
efficiently. Our experiments and evaluations show the effectiveness of our
proposed metric of automatic weight generation for properties in an instance
matching technique.
|
1502.00996 | Brian Thomas | Brian Thomas, Tim Jenness, Frossie Economou, Perry Greenfield, Paul
Hirst, David S. Berry, Erik Bray, Norman Gray, Demitri Muna, James Turner,
Miguel de Val-Borro, Juande Santander-Vela, David Shupe, John Good, G. Bruce
Berriman, Slava Kitaeff, Jonathan Fay, Omar Laurino, Anastasia Alexov, Walter
Landry, Joe Masters, Adam Brazier, Reinhold Schaaf, Kevin Edwards, Russell O.
Redman, Thomas R. Marsh, Ole Streicher, Pat Norris, Sergio Pascual, Matthew
Davie, Michael Droettboom, Thomas Robitaille, Riccardo Campana, Alex Hagen,
Paul Hartogh, Dominik Klaes, Matthew W. Craig, Derek Homeier | Learning from FITS: Limitations in use in modern astronomical research | null | null | 10.1016/j.ascom.2015.01.009 | null | astro-ph.IM cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Flexible Image Transport System (FITS) standard has been a great boon to
astronomy, allowing observatories, scientists and the public to exchange
astronomical information easily. The FITS standard, however, is showing its
age. Developed in the late 1970s, the FITS authors made a number of
implementation choices that, while common at the time, are now seen to limit
its utility with modern data. The authors of the FITS standard could not
anticipate the challenges which we are facing today in astronomical computing.
Difficulties we now face include, but are not limited to, addressing the need
to handle an expanded range of specialized data product types (data models),
being more conducive to the networked exchange and storage of data, handling
very large datasets, and capturing significantly more complex metadata and data
relationships.
There are members of the community today who find some or all of these
limitations unworkable, and have decided to move ahead with storing data in
other formats. If this fragmentation continues, we risk abandoning the
advantages of broad interoperability, and ready archivability, that the FITS
format provides for astronomy. In this paper we detail some selected important
problems which exist within the FITS standard today. These problems may provide
insight into deeper underlying issues which reside in the format and we provide
a discussion of some lessons learned. It is not our intention here to prescribe
specific remedies to these issues; rather, it is to call attention of the FITS
and greater astronomical computing communities to these problems in the hope
that it will spur action to address them.
| [
{
"version": "v1",
"created": "Tue, 3 Feb 2015 20:27:29 GMT"
},
{
"version": "v2",
"created": "Tue, 10 Feb 2015 21:35:57 GMT"
}
] | 2015-02-12T00:00:00 | [
[
"Thomas",
"Brian",
""
],
[
"Jenness",
"Tim",
""
],
[
"Economou",
"Frossie",
""
],
[
"Greenfield",
"Perry",
""
],
[
"Hirst",
"Paul",
""
],
[
"Berry",
"David S.",
""
],
[
"Bray",
"Erik",
""
],
[
"Gray",
"Norman",
""
],
[
"Muna",
"Demitri",
""
],
[
"Turner",
"James",
""
],
[
"de Val-Borro",
"Miguel",
""
],
[
"Santander-Vela",
"Juande",
""
],
[
"Shupe",
"David",
""
],
[
"Good",
"John",
""
],
[
"Berriman",
"G. Bruce",
""
],
[
"Kitaeff",
"Slava",
""
],
[
"Fay",
"Jonathan",
""
],
[
"Laurino",
"Omar",
""
],
[
"Alexov",
"Anastasia",
""
],
[
"Landry",
"Walter",
""
],
[
"Masters",
"Joe",
""
],
[
"Brazier",
"Adam",
""
],
[
"Schaaf",
"Reinhold",
""
],
[
"Edwards",
"Kevin",
""
],
[
"Redman",
"Russell O.",
""
],
[
"Marsh",
"Thomas R.",
""
],
[
"Streicher",
"Ole",
""
],
[
"Norris",
"Pat",
""
],
[
"Pascual",
"Sergio",
""
],
[
"Davie",
"Matthew",
""
],
[
"Droettboom",
"Michael",
""
],
[
"Robitaille",
"Thomas",
""
],
[
"Campana",
"Riccardo",
""
],
[
"Hagen",
"Alex",
""
],
[
"Hartogh",
"Paul",
""
],
[
"Klaes",
"Dominik",
""
],
[
"Craig",
"Matthew W.",
""
],
[
"Homeier",
"Derek",
""
]
] | TITLE: Learning from FITS: Limitations in use in modern astronomical research
ABSTRACT: The Flexible Image Transport System (FITS) standard has been a great boon to
astronomy, allowing observatories, scientists and the public to exchange
astronomical information easily. The FITS standard, however, is showing its
age. Developed in the late 1970s, the FITS authors made a number of
implementation choices that, while common at the time, are now seen to limit
its utility with modern data. The authors of the FITS standard could not
anticipate the challenges which we are facing today in astronomical computing.
Difficulties we now face include, but are not limited to, addressing the need
to handle an expanded range of specialized data product types (data models),
being more conducive to the networked exchange and storage of data, handling
very large datasets, and capturing significantly more complex metadata and data
relationships.
There are members of the community today who find some or all of these
limitations unworkable, and have decided to move ahead with storing data in
other formats. If this fragmentation continues, we risk abandoning the
advantages of broad interoperability, and ready archivability, that the FITS
format provides for astronomy. In this paper we detail some selected important
problems which exist within the FITS standard today. These problems may provide
insight into deeper underlying issues which reside in the format and we provide
a discussion of some lessons learned. It is not our intention here to prescribe
specific remedies to these issues; rather, it is to call attention of the FITS
and greater astronomical computing communities to these problems in the hope
that it will spur action to address them.
|
1502.02772 | Kean Lau Hong | Kean Hong Lau, Yong Haur Tay, Fook Loong Lo | A HMAX with LLC for visual recognition | 10 pages, 3 figures, 2 tables, 23 references | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Today's high performance deep artificial neural networks (ANNs) rely heavily
on parameter optimization, which is sequential in nature and even with a
powerful GPU, would have taken weeks to train them up for solving challenging
tasks [22]. HMAX [17] has demonstrated that a simple high performing network
could be obtained without heavy optimization. In this paper, we had improved on
the existing best HMAX neural network [12] in terms of structural simplicity
and performance. Our design replaces the L1 minimization sparse coding (SC)
with a locality-constrained linear coding (LLC) [20] which has a lower
computational demand. We also put the simple orientation filter bank back into
the front layer of the network replacing PCA. Our system's performance has
improved over the existing architecture and reached 79.0% on the challenging
Caltech-101 [7] dataset, which is state-of-the-art for ANNs (without transfer
learning). From our empirical data, the main contributors to our system's
performance include an introduction of partial signal whitening, a spot
detector, and a spatial pyramid matching (SPM) [14] layer.
| [
{
"version": "v1",
"created": "Tue, 10 Feb 2015 04:01:43 GMT"
},
{
"version": "v2",
"created": "Wed, 11 Feb 2015 04:40:20 GMT"
}
] | 2015-02-12T00:00:00 | [
[
"Lau",
"Kean Hong",
""
],
[
"Tay",
"Yong Haur",
""
],
[
"Lo",
"Fook Loong",
""
]
] | TITLE: A HMAX with LLC for visual recognition
ABSTRACT: Today's high performance deep artificial neural networks (ANNs) rely heavily
on parameter optimization, which is sequential in nature and even with a
powerful GPU, would have taken weeks to train them up for solving challenging
tasks [22]. HMAX [17] has demonstrated that a simple high performing network
could be obtained without heavy optimization. In this paper, we had improved on
the existing best HMAX neural network [12] in terms of structural simplicity
and performance. Our design replaces the L1 minimization sparse coding (SC)
with a locality-constrained linear coding (LLC) [20] which has a lower
computational demand. We also put the simple orientation filter bank back into
the front layer of the network replacing PCA. Our system's performance has
improved over the existing architecture and reached 79.0% on the challenging
Caltech-101 [7] dataset, which is state-of-the-art for ANNs (without transfer
learning). From our empirical data, the main contributors to our system's
performance include an introduction of partial signal whitening, a spot
detector, and a spatial pyramid matching (SPM) [14] layer.
|
1502.03406 | Adeline Decuyper | Vincent D. Blondel, Adeline Decuyper, Gautier Krings | A survey of results on mobile phone datasets analysis | null | null | null | null | physics.soc-ph cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we review some advances made recently in the study of mobile
phone datasets. This area of research has emerged a decade ago, with the
increasing availability of large-scale anonymized datasets, and has grown into
a stand-alone topic. We will survey the contributions made so far on the social
networks that can be constructed with such data, the study of personal
mobility, geographical partitioning, urban planning, and help towards
development as well as security and privacy issues.
| [
{
"version": "v1",
"created": "Wed, 11 Feb 2015 18:58:12 GMT"
}
] | 2015-02-12T00:00:00 | [
[
"Blondel",
"Vincent D.",
""
],
[
"Decuyper",
"Adeline",
""
],
[
"Krings",
"Gautier",
""
]
] | TITLE: A survey of results on mobile phone datasets analysis
ABSTRACT: In this paper, we review some advances made recently in the study of mobile
phone datasets. This area of research has emerged a decade ago, with the
increasing availability of large-scale anonymized datasets, and has grown into
a stand-alone topic. We will survey the contributions made so far on the social
networks that can be constructed with such data, the study of personal
mobility, geographical partitioning, urban planning, and help towards
development as well as security and privacy issues.
|
1502.03409 | Karl Ni | Karl Ni, Roger Pearce, Kofi Boakye, Brian Van Essen, Damian Borth,
Barry Chen, Eric Wang | Large-Scale Deep Learning on the YFCC100M Dataset | null | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/publicdomain/ | We present a work-in-progress snapshot of learning with a 15 billion
parameter deep learning network on HPC architectures applied to the largest
publicly available natural image and video dataset released to-date. Recent
advancements in unsupervised deep neural networks suggest that scaling up such
networks in both model and training dataset size can yield significant
improvements in the learning of concepts at the highest layers. We train our
three-layer deep neural network on the Yahoo! Flickr Creative Commons 100M
dataset. The dataset comprises approximately 99.2 million images and 800,000
user-created videos from Yahoo's Flickr image and video sharing platform.
Training of our network takes eight days on 98 GPU nodes at the High
Performance Computing Center at Lawrence Livermore National Laboratory.
Encouraging preliminary results and future research directions are presented
and discussed.
| [
{
"version": "v1",
"created": "Wed, 11 Feb 2015 19:24:36 GMT"
}
] | 2015-02-12T00:00:00 | [
[
"Ni",
"Karl",
""
],
[
"Pearce",
"Roger",
""
],
[
"Boakye",
"Kofi",
""
],
[
"Van Essen",
"Brian",
""
],
[
"Borth",
"Damian",
""
],
[
"Chen",
"Barry",
""
],
[
"Wang",
"Eric",
""
]
] | TITLE: Large-Scale Deep Learning on the YFCC100M Dataset
ABSTRACT: We present a work-in-progress snapshot of learning with a 15 billion
parameter deep learning network on HPC architectures applied to the largest
publicly available natural image and video dataset released to-date. Recent
advancements in unsupervised deep neural networks suggest that scaling up such
networks in both model and training dataset size can yield significant
improvements in the learning of concepts at the highest layers. We train our
three-layer deep neural network on the Yahoo! Flickr Creative Commons 100M
dataset. The dataset comprises approximately 99.2 million images and 800,000
user-created videos from Yahoo's Flickr image and video sharing platform.
Training of our network takes eight days on 98 GPU nodes at the High
Performance Computing Center at Lawrence Livermore National Laboratory.
Encouraging preliminary results and future research directions are presented
and discussed.
|
1404.7514 | Mihai Alexandru Petrovici | Mihai A. Petrovici, Bernhard Vogginger, Paul M\"uller, Oliver
Breitwieser, Mikael Lundqvist, Lyle Muller, Matthias Ehrlich, Alain Destexhe,
Anders Lansner, Ren\'e Sch\"uffny, Johannes Schemmel, Karlheinz Meier | Characterization and Compensation of Network-Level Anomalies in
Mixed-Signal Neuromorphic Modeling Platforms | null | PLOS ONE, October 10th 2014 | 10.1371/journal.pone.0108590 | null | q-bio.NC cond-mat.dis-nn cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Advancing the size and complexity of neural network models leads to an ever
increasing demand for computational resources for their simulation.
Neuromorphic devices offer a number of advantages over conventional computing
architectures, such as high emulation speed or low power consumption, but this
usually comes at the price of reduced configurability and precision. In this
article, we investigate the consequences of several such factors that are
common to neuromorphic devices, more specifically limited hardware resources,
limited parameter configurability and parameter variations. Our final aim is to
provide an array of methods for coping with such inevitable distortion
mechanisms. As a platform for testing our proposed strategies, we use an
executable system specification (ESS) of the BrainScaleS neuromorphic system,
which has been designed as a universal emulation back-end for neuroscientific
modeling. We address the most essential limitations of this device in detail
and study their effects on three prototypical benchmark network models within a
well-defined, systematic workflow. For each network model, we start by defining
quantifiable functionality measures by which we then assess the effects of
typical hardware-specific distortion mechanisms, both in idealized software
simulations and on the ESS. For those effects that cause unacceptable
deviations from the original network dynamics, we suggest generic compensation
mechanisms and demonstrate their effectiveness. Both the suggested workflow and
the investigated compensation mechanisms are largely back-end independent and
do not require additional hardware configurability beyond the one required to
emulate the benchmark networks in the first place. We hereby provide a generic
methodological environment for configurable neuromorphic devices that are
targeted at emulating large-scale, functional neural networks.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2014 20:20:13 GMT"
},
{
"version": "v2",
"created": "Tue, 10 Feb 2015 14:05:53 GMT"
}
] | 2015-02-11T00:00:00 | [
[
"Petrovici",
"Mihai A.",
""
],
[
"Vogginger",
"Bernhard",
""
],
[
"Müller",
"Paul",
""
],
[
"Breitwieser",
"Oliver",
""
],
[
"Lundqvist",
"Mikael",
""
],
[
"Muller",
"Lyle",
""
],
[
"Ehrlich",
"Matthias",
""
],
[
"Destexhe",
"Alain",
""
],
[
"Lansner",
"Anders",
""
],
[
"Schüffny",
"René",
""
],
[
"Schemmel",
"Johannes",
""
],
[
"Meier",
"Karlheinz",
""
]
] | TITLE: Characterization and Compensation of Network-Level Anomalies in
Mixed-Signal Neuromorphic Modeling Platforms
ABSTRACT: Advancing the size and complexity of neural network models leads to an ever
increasing demand for computational resources for their simulation.
Neuromorphic devices offer a number of advantages over conventional computing
architectures, such as high emulation speed or low power consumption, but this
usually comes at the price of reduced configurability and precision. In this
article, we investigate the consequences of several such factors that are
common to neuromorphic devices, more specifically limited hardware resources,
limited parameter configurability and parameter variations. Our final aim is to
provide an array of methods for coping with such inevitable distortion
mechanisms. As a platform for testing our proposed strategies, we use an
executable system specification (ESS) of the BrainScaleS neuromorphic system,
which has been designed as a universal emulation back-end for neuroscientific
modeling. We address the most essential limitations of this device in detail
and study their effects on three prototypical benchmark network models within a
well-defined, systematic workflow. For each network model, we start by defining
quantifiable functionality measures by which we then assess the effects of
typical hardware-specific distortion mechanisms, both in idealized software
simulations and on the ESS. For those effects that cause unacceptable
deviations from the original network dynamics, we suggest generic compensation
mechanisms and demonstrate their effectiveness. Both the suggested workflow and
the investigated compensation mechanisms are largely back-end independent and
do not require additional hardware configurability beyond the one required to
emulate the benchmark networks in the first place. We hereby provide a generic
methodological environment for configurable neuromorphic devices that are
targeted at emulating large-scale, functional neural networks.
|
1502.02761 | Yujia Li | Yujia Li, Kevin Swersky and Richard Zemel | Generative Moment Matching Networks | null | null | null | null | cs.LG cs.AI stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the problem of learning deep generative models from data. We
formulate a method that generates an independent sample via a single
feedforward pass through a multilayer perceptron, as in the recently proposed
generative adversarial networks (Goodfellow et al., 2014). Training a
generative adversarial network, however, requires careful optimization of a
difficult minimax program. Instead, we utilize a technique from statistical
hypothesis testing known as maximum mean discrepancy (MMD), which leads to a
simple objective that can be interpreted as matching all orders of statistics
between a dataset and samples from the model, and can be trained by
backpropagation. We further boost the performance of this approach by combining
our generative network with an auto-encoder network, using MMD to learn to
generate codes that can then be decoded to produce samples. We show that the
combination of these techniques yields excellent generative models compared to
baseline approaches as measured on MNIST and the Toronto Face Database.
| [
{
"version": "v1",
"created": "Tue, 10 Feb 2015 02:54:58 GMT"
}
] | 2015-02-11T00:00:00 | [
[
"Li",
"Yujia",
""
],
[
"Swersky",
"Kevin",
""
],
[
"Zemel",
"Richard",
""
]
] | TITLE: Generative Moment Matching Networks
ABSTRACT: We consider the problem of learning deep generative models from data. We
formulate a method that generates an independent sample via a single
feedforward pass through a multilayer perceptron, as in the recently proposed
generative adversarial networks (Goodfellow et al., 2014). Training a
generative adversarial network, however, requires careful optimization of a
difficult minimax program. Instead, we utilize a technique from statistical
hypothesis testing known as maximum mean discrepancy (MMD), which leads to a
simple objective that can be interpreted as matching all orders of statistics
between a dataset and samples from the model, and can be trained by
backpropagation. We further boost the performance of this approach by combining
our generative network with an auto-encoder network, using MMD to learn to
generate codes that can then be decoded to produce samples. We show that the
combination of these techniques yields excellent generative models compared to
baseline approaches as measured on MNIST and the Toronto Face Database.
|
1502.02821 | Paul Doyle Dr. | Paul Doyle | Building a scalable global data processing pipeline for large
astronomical photometric datasets | PhD Thesis, Dublin Institute of Technology | null | null | null | astro-ph.IM cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Astronomical photometry is the science of measuring the flux of a celestial
object. Since its introduction, the CCD has been the principle method of
measuring flux to calculate the apparent magnitude of an object. Each CCD image
taken must go through a process of cleaning and calibration prior to its use.
As the number of research telescopes increases the overall computing resources
required for image processing also increases. Existing processing techniques
are primarily sequential in nature, requiring increasingly powerful servers,
faster disks and faster networks to process data. Existing High Performance
Computing solutions involving high capacity data centres are complex in design
and expensive to maintain, while providing resources primarily to high profile
science projects. This research describes three distributed pipeline
architectures, a virtualised cloud based IRAF, the Astronomical Compute Node
(ACN), a private cloud based pipeline, and NIMBUS, a globally distributed
system. The ACN pipeline processed data at a rate of 4 Terabytes per day
demonstrating data compression and upload to a central cloud storage service at
a rate faster than data generation. The primary contribution of this research
is NIMBUS, which is rapidly scalable, resilient to failure and capable of
processing CCD image data at a rate of hundreds of Terabytes per day. This
pipeline is implemented using a decentralised web queue to control the
compression of data, uploading of data to distributed web servers, and creating
web messages to identify the location of the data. Using distributed web queue
messages, images are downloaded by computing resources distributed around the
globe. Rigorous experimental evidence is presented verifying the horizontal
scalability of the system which has demonstrated a processing rate of 192
Terabytes per day with clear indications that higher processing rates are
possible.
| [
{
"version": "v1",
"created": "Tue, 10 Feb 2015 09:27:58 GMT"
}
] | 2015-02-11T00:00:00 | [
[
"Doyle",
"Paul",
""
]
] | TITLE: Building a scalable global data processing pipeline for large
astronomical photometric datasets
ABSTRACT: Astronomical photometry is the science of measuring the flux of a celestial
object. Since its introduction, the CCD has been the principle method of
measuring flux to calculate the apparent magnitude of an object. Each CCD image
taken must go through a process of cleaning and calibration prior to its use.
As the number of research telescopes increases the overall computing resources
required for image processing also increases. Existing processing techniques
are primarily sequential in nature, requiring increasingly powerful servers,
faster disks and faster networks to process data. Existing High Performance
Computing solutions involving high capacity data centres are complex in design
and expensive to maintain, while providing resources primarily to high profile
science projects. This research describes three distributed pipeline
architectures, a virtualised cloud based IRAF, the Astronomical Compute Node
(ACN), a private cloud based pipeline, and NIMBUS, a globally distributed
system. The ACN pipeline processed data at a rate of 4 Terabytes per day
demonstrating data compression and upload to a central cloud storage service at
a rate faster than data generation. The primary contribution of this research
is NIMBUS, which is rapidly scalable, resilient to failure and capable of
processing CCD image data at a rate of hundreds of Terabytes per day. This
pipeline is implemented using a decentralised web queue to control the
compression of data, uploading of data to distributed web servers, and creating
web messages to identify the location of the data. Using distributed web queue
messages, images are downloaded by computing resources distributed around the
globe. Rigorous experimental evidence is presented verifying the horizontal
scalability of the system which has demonstrated a processing rate of 192
Terabytes per day with clear indications that higher processing rates are
possible.
|
1502.02852 | Daniel Gregorek | Daniel Gregorek, Robert Schmidt, Alberto Garcia-Ortiz | Transaction Level Analysis for a Clustered and Hardware-Enhanced Task
Manager on Homogeneous Many-Core Systems | Presented at HIP3ES, 2015 (arXiv: 1501.03064) | null | null | HIP3ES/2015/03 | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The increasing parallelism of many-core systems demands for efficient
strategies for the run-time system management. Due to the large number of cores
the management overhead has a rising impact to the overall system performance.
This work analyzes a clustered infrastructure of dedicated hardware nodes to
manage a homogeneous many-core system. The hardware nodes implement a message
passing protocol and perform the task mapping and synchronization at run-time.
To make meaningful mapping decisions, the global management nodes employ a
workload status communication mechanism.
This paper discusses the design-space of the dedicated infrastructure by
means of task mapping use-cases and a parallel benchmark including
application-interference. We evaluate the architecture in terms of application
speedup and analyze the mechanism for the status communication. A comparison
versus centralized and fully-distributed configurations demonstrates the
reduction of the computation and communication management overhead for our
approach.
| [
{
"version": "v1",
"created": "Tue, 10 Feb 2015 10:56:50 GMT"
}
] | 2015-02-11T00:00:00 | [
[
"Gregorek",
"Daniel",
""
],
[
"Schmidt",
"Robert",
""
],
[
"Garcia-Ortiz",
"Alberto",
""
]
] | TITLE: Transaction Level Analysis for a Clustered and Hardware-Enhanced Task
Manager on Homogeneous Many-Core Systems
ABSTRACT: The increasing parallelism of many-core systems demands for efficient
strategies for the run-time system management. Due to the large number of cores
the management overhead has a rising impact to the overall system performance.
This work analyzes a clustered infrastructure of dedicated hardware nodes to
manage a homogeneous many-core system. The hardware nodes implement a message
passing protocol and perform the task mapping and synchronization at run-time.
To make meaningful mapping decisions, the global management nodes employ a
workload status communication mechanism.
This paper discusses the design-space of the dedicated infrastructure by
means of task mapping use-cases and a parallel benchmark including
application-interference. We evaluate the architecture in terms of application
speedup and analyze the mechanism for the status communication. A comparison
versus centralized and fully-distributed configurations demonstrates the
reduction of the computation and communication management overhead for our
approach.
|
cs/0604054 | Maria Paola Bonacina | Alessandro Armando, Maria Paola Bonacina, Silvio Ranise, Stephan
Schulz | New results on rewrite-based satisfiability procedures | To appear in the ACM Transactions on Computational Logic, 49 pages | ACM Transactions on Computational Logic, 10(1):129-179, January
2009 | 10.1145/1459010.1459014 | RR 36/2005 | cs.AI cs.LO | null | Program analysis and verification require decision procedures to reason on
theories of data structures. Many problems can be reduced to the satisfiability
of sets of ground literals in theory T. If a sound and complete inference
system for first-order logic is guaranteed to terminate on T-satisfiability
problems, any theorem-proving strategy with that system and a fair search plan
is a T-satisfiability procedure. We prove termination of a rewrite-based
first-order engine on the theories of records, integer offsets, integer offsets
modulo and lists. We give a modularity theorem stating sufficient conditions
for termination on a combinations of theories, given termination on each. The
above theories, as well as others, satisfy these conditions. We introduce
several sets of benchmarks on these theories and their combinations, including
both parametric synthetic benchmarks to test scalability, and real-world
problems to test performances on huge sets of literals. We compare the
rewrite-based theorem prover E with the validity checkers CVC and CVC Lite.
Contrary to the folklore that a general-purpose prover cannot compete with
reasoners with built-in theories, the experiments are overall favorable to the
theorem prover, showing that not only the rewriting approach is elegant and
conceptually simple, but has important practical implications.
| [
{
"version": "v1",
"created": "Wed, 12 Apr 2006 19:53:24 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Nov 2006 19:54:27 GMT"
},
{
"version": "v3",
"created": "Wed, 6 Jun 2007 17:15:23 GMT"
},
{
"version": "v4",
"created": "Sun, 1 Jun 2008 02:02:27 GMT"
}
] | 2015-02-11T00:00:00 | [
[
"Armando",
"Alessandro",
""
],
[
"Bonacina",
"Maria Paola",
""
],
[
"Ranise",
"Silvio",
""
],
[
"Schulz",
"Stephan",
""
]
] | TITLE: New results on rewrite-based satisfiability procedures
ABSTRACT: Program analysis and verification require decision procedures to reason on
theories of data structures. Many problems can be reduced to the satisfiability
of sets of ground literals in theory T. If a sound and complete inference
system for first-order logic is guaranteed to terminate on T-satisfiability
problems, any theorem-proving strategy with that system and a fair search plan
is a T-satisfiability procedure. We prove termination of a rewrite-based
first-order engine on the theories of records, integer offsets, integer offsets
modulo and lists. We give a modularity theorem stating sufficient conditions
for termination on a combinations of theories, given termination on each. The
above theories, as well as others, satisfy these conditions. We introduce
several sets of benchmarks on these theories and their combinations, including
both parametric synthetic benchmarks to test scalability, and real-world
problems to test performances on huge sets of literals. We compare the
rewrite-based theorem prover E with the validity checkers CVC and CVC Lite.
Contrary to the folklore that a general-purpose prover cannot compete with
reasoners with built-in theories, the experiments are overall favorable to the
theorem prover, showing that not only the rewriting approach is elegant and
conceptually simple, but has important practical implications.
|
1407.8186 | Xiaoting Zhao | Xiaoting Zhao, Peter I. Frazier | Exploration vs. Exploitation in the Information Filtering Problem | 36 pages, 5 figures | null | null | null | math.OC cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider information filtering, in which we face a stream of items too
voluminous to process by hand (e.g., scientific articles, blog posts, emails),
and must rely on a computer system to automatically filter out irrelevant
items. Such systems face the exploration vs. exploitation tradeoff, in which it
may be beneficial to present an item despite a low probability of relevance,
just to learn about future items with similar content. We present a Bayesian
sequential decision-making model of this problem, show how it may be solved to
optimality using a decomposition to a collection of two-armed bandit problems,
and show structural results for the optimal policy. We show that the resulting
method is especially useful when facing the cold start problem, i.e., when
filtering items for new users without a long history of past interactions. We
then present an application of this information filtering method to a
historical dataset from the arXiv.org repository of scientific articles.
| [
{
"version": "v1",
"created": "Wed, 30 Jul 2014 20:00:11 GMT"
},
{
"version": "v2",
"created": "Tue, 2 Sep 2014 20:01:00 GMT"
},
{
"version": "v3",
"created": "Sun, 8 Feb 2015 21:03:08 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Zhao",
"Xiaoting",
""
],
[
"Frazier",
"Peter I.",
""
]
] | TITLE: Exploration vs. Exploitation in the Information Filtering Problem
ABSTRACT: We consider information filtering, in which we face a stream of items too
voluminous to process by hand (e.g., scientific articles, blog posts, emails),
and must rely on a computer system to automatically filter out irrelevant
items. Such systems face the exploration vs. exploitation tradeoff, in which it
may be beneficial to present an item despite a low probability of relevance,
just to learn about future items with similar content. We present a Bayesian
sequential decision-making model of this problem, show how it may be solved to
optimality using a decomposition to a collection of two-armed bandit problems,
and show structural results for the optimal policy. We show that the resulting
method is especially useful when facing the cold start problem, i.e., when
filtering items for new users without a long history of past interactions. We
then present an application of this information filtering method to a
historical dataset from the arXiv.org repository of scientific articles.
|
1412.4446 | Pascal Germain | Hana Ajakan, Pascal Germain, Hugo Larochelle, Fran\c{c}ois Laviolette,
Mario Marchand | Domain-Adversarial Neural Networks | The first version of this paper was accepted at the "Second Workshop
on Transfer and Multi-Task Learning: Theory meets Practice" (NIPS 2014,
Montreal, Canada). See: https://sites.google.com/site/multitaskwsnips2014/ | null | null | null | stat.ML cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a new representation learning algorithm suited to the context of
domain adaptation, in which data at training and test time come from similar
but different distributions. Our algorithm is directly inspired by theory on
domain adaptation suggesting that, for effective domain transfer to be
achieved, predictions must be made based on a data representation that cannot
discriminate between the training (source) and test (target) domains. We
propose a training objective that implements this idea in the context of a
neural network, whose hidden layer is trained to be predictive of the
classification task, but uninformative as to the domain of the input. Our
experiments on a sentiment analysis classification benchmark, where the target
domain data available at training time is unlabeled, show that our neural
network for domain adaption algorithm has better performance than either a
standard neural network or an SVM, even if trained on input features extracted
with the state-of-the-art marginalized stacked denoising autoencoders of Chen
et al. (2012).
| [
{
"version": "v1",
"created": "Mon, 15 Dec 2014 02:16:07 GMT"
},
{
"version": "v2",
"created": "Mon, 9 Feb 2015 17:52:03 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Ajakan",
"Hana",
""
],
[
"Germain",
"Pascal",
""
],
[
"Larochelle",
"Hugo",
""
],
[
"Laviolette",
"François",
""
],
[
"Marchand",
"Mario",
""
]
] | TITLE: Domain-Adversarial Neural Networks
ABSTRACT: We introduce a new representation learning algorithm suited to the context of
domain adaptation, in which data at training and test time come from similar
but different distributions. Our algorithm is directly inspired by theory on
domain adaptation suggesting that, for effective domain transfer to be
achieved, predictions must be made based on a data representation that cannot
discriminate between the training (source) and test (target) domains. We
propose a training objective that implements this idea in the context of a
neural network, whose hidden layer is trained to be predictive of the
classification task, but uninformative as to the domain of the input. Our
experiments on a sentiment analysis classification benchmark, where the target
domain data available at training time is unlabeled, show that our neural
network for domain adaption algorithm has better performance than either a
standard neural network or an SVM, even if trained on input features extracted
with the state-of-the-art marginalized stacked denoising autoencoders of Chen
et al. (2012).
|
1501.03326 | Heiko Strathmann | Heiko Strathmann, Dino Sejdinovic, Mark Girolami | Unbiased Bayes for Big Data: Paths of Partial Posteriors | 18 pages, 10 figures | null | null | null | stat.ML cs.LG stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A key quantity of interest in Bayesian inference are expectations of
functions with respect to a posterior distribution. Markov Chain Monte Carlo is
a fundamental tool to consistently compute these expectations via averaging
samples drawn from an approximate posterior. However, its feasibility is being
challenged in the era of so called Big Data as all data needs to be processed
in every iteration. Realising that such simulation is an unnecessarily hard
problem if the goal is estimation, we construct a computationally scalable
methodology that allows unbiased estimation of the required expectations --
without explicit simulation from the full posterior. The scheme's variance is
finite by construction and straightforward to control, leading to algorithms
that are provably unbiased and naturally arrive at a desired error tolerance.
This is achieved at an average computational complexity that is sub-linear in
the size of the dataset and its free parameters are easy to tune. We
demonstrate the utility and generality of the methodology on a range of common
statistical models applied to large-scale benchmark and real-world datasets.
| [
{
"version": "v1",
"created": "Wed, 14 Jan 2015 12:15:14 GMT"
},
{
"version": "v2",
"created": "Mon, 9 Feb 2015 16:21:20 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Strathmann",
"Heiko",
""
],
[
"Sejdinovic",
"Dino",
""
],
[
"Girolami",
"Mark",
""
]
] | TITLE: Unbiased Bayes for Big Data: Paths of Partial Posteriors
ABSTRACT: A key quantity of interest in Bayesian inference are expectations of
functions with respect to a posterior distribution. Markov Chain Monte Carlo is
a fundamental tool to consistently compute these expectations via averaging
samples drawn from an approximate posterior. However, its feasibility is being
challenged in the era of so called Big Data as all data needs to be processed
in every iteration. Realising that such simulation is an unnecessarily hard
problem if the goal is estimation, we construct a computationally scalable
methodology that allows unbiased estimation of the required expectations --
without explicit simulation from the full posterior. The scheme's variance is
finite by construction and straightforward to control, leading to algorithms
that are provably unbiased and naturally arrive at a desired error tolerance.
This is achieved at an average computational complexity that is sub-linear in
the size of the dataset and its free parameters are easy to tune. We
demonstrate the utility and generality of the methodology on a range of common
statistical models applied to large-scale benchmark and real-world datasets.
|
1502.00558 | Marcelo Cicconet | Marcelo Cicconet, Davi Geiger, and Michael Werman | Complex-Valued Hough Transforms for Circles | The paper has been withdrawn since the authors concluded a more
comprehensive study on the choice of parameters needs to be performed | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper advocates the use of complex variables to represent votes in the
Hough transform for circle detection. Replacing the positive numbers
classically used in the parameter space of the Hough transforms by complex
numbers allows cancellation effects when adding up the votes. Cancellation and
the computation of shape likelihood via a complex number's magnitude square
lead to more robust solutions than the "classic" algorithms, as shown by
computational experiments on synthetic and real datasets.
| [
{
"version": "v1",
"created": "Mon, 2 Feb 2015 17:22:26 GMT"
},
{
"version": "v2",
"created": "Mon, 9 Feb 2015 19:38:58 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Cicconet",
"Marcelo",
""
],
[
"Geiger",
"Davi",
""
],
[
"Werman",
"Michael",
""
]
] | TITLE: Complex-Valued Hough Transforms for Circles
ABSTRACT: This paper advocates the use of complex variables to represent votes in the
Hough transform for circle detection. Replacing the positive numbers
classically used in the parameter space of the Hough transforms by complex
numbers allows cancellation effects when adding up the votes. Cancellation and
the computation of shape likelihood via a complex number's magnitude square
lead to more robust solutions than the "classic" algorithms, as shown by
computational experiments on synthetic and real datasets.
|
1502.02072 | Bharath Ramsundar | Bharath Ramsundar, Steven Kearnes, Patrick Riley, Dale Webster, David
Konerding, Vijay Pande | Massively Multitask Networks for Drug Discovery | Preliminary work. Under review by the International Conference on
Machine Learning (ICML) | null | null | null | stat.ML cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Massively multitask neural architectures provide a learning framework for
drug discovery that synthesizes information from many distinct biological
sources. To train these architectures at scale, we gather large amounts of data
from public sources to create a dataset of nearly 40 million measurements
across more than 200 biological targets. We investigate several aspects of the
multitask framework by performing a series of empirical studies and obtain some
interesting results: (1) massively multitask networks obtain predictive
accuracies significantly better than single-task methods, (2) the predictive
power of multitask networks improves as additional tasks and data are added,
(3) the total amount of data and the total number of tasks both contribute
significantly to multitask improvement, and (4) multitask networks afford
limited transferability to tasks not in the training set. Our results
underscore the need for greater data sharing and further algorithmic innovation
to accelerate the drug discovery process.
| [
{
"version": "v1",
"created": "Fri, 6 Feb 2015 23:04:01 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Ramsundar",
"Bharath",
""
],
[
"Kearnes",
"Steven",
""
],
[
"Riley",
"Patrick",
""
],
[
"Webster",
"Dale",
""
],
[
"Konerding",
"David",
""
],
[
"Pande",
"Vijay",
""
]
] | TITLE: Massively Multitask Networks for Drug Discovery
ABSTRACT: Massively multitask neural architectures provide a learning framework for
drug discovery that synthesizes information from many distinct biological
sources. To train these architectures at scale, we gather large amounts of data
from public sources to create a dataset of nearly 40 million measurements
across more than 200 biological targets. We investigate several aspects of the
multitask framework by performing a series of empirical studies and obtain some
interesting results: (1) massively multitask networks obtain predictive
accuracies significantly better than single-task methods, (2) the predictive
power of multitask networks improves as additional tasks and data are added,
(3) the total amount of data and the total number of tasks both contribute
significantly to multitask improvement, and (4) multitask networks afford
limited transferability to tasks not in the training set. Our results
underscore the need for greater data sharing and further algorithmic innovation
to accelerate the drug discovery process.
|
1502.02134 | Derrick Wing Kwan Ng | Derrick Wing Kwan Ng, Yongpeng Wu, and Robert Schober | Power Efficient Resource Allocation for Full-Duplex Radio Distributed
Antenna Networks | Submitted for possible journal publication | null | null | null | cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we study the resource allocation algorithm design for
distributed antenna multiuser networks with full-duplex (FD) radio base
stations (BSs) which enable simultaneous uplink and downlink communications.
The considered resource allocation algorithm design is formulated as an
optimization problem taking into account the antenna circuit power consumption
of the BSs and the quality of service (QoS) requirements of both uplink and
downlink users. We minimize the total network power consumption by jointly
optimizing the downlink beamformer, the uplink transmit power, and the antenna
selection. To overcome the intractability of the resulting problem, we
reformulate it as an optimization problem with decoupled binary selection
variables and non-convex constraints. The reformulated problem facilitates the
design of an iterative resource allocation algorithm which obtains an optimal
solution based on the generalized Bender's decomposition (GBD) and serves as a
benchmark scheme. Furthermore, to strike a balance between computational
complexity and system performance, a suboptimal algorithm with polynomial time
complexity is proposed. Simulation results illustrate that the proposed GBD
based iterative algorithm converges to the global optimal solution and the
suboptimal algorithm achieves a close-to-optimal performance. Our results also
demonstrate the trade-off between power efficiency and the number of active
transmit antennas when the circuit power consumption is taken into account. In
particular, activating an exceedingly large number of antennas may not be a
power efficient solution for reducing the total system power consumption. In
addition, our results reveal that FD systems facilitate significant power
savings compared to traditional half-duplex systems, despite the non-negligible
self-interference.
| [
{
"version": "v1",
"created": "Sat, 7 Feb 2015 12:50:13 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Ng",
"Derrick Wing Kwan",
""
],
[
"Wu",
"Yongpeng",
""
],
[
"Schober",
"Robert",
""
]
] | TITLE: Power Efficient Resource Allocation for Full-Duplex Radio Distributed
Antenna Networks
ABSTRACT: In this paper, we study the resource allocation algorithm design for
distributed antenna multiuser networks with full-duplex (FD) radio base
stations (BSs) which enable simultaneous uplink and downlink communications.
The considered resource allocation algorithm design is formulated as an
optimization problem taking into account the antenna circuit power consumption
of the BSs and the quality of service (QoS) requirements of both uplink and
downlink users. We minimize the total network power consumption by jointly
optimizing the downlink beamformer, the uplink transmit power, and the antenna
selection. To overcome the intractability of the resulting problem, we
reformulate it as an optimization problem with decoupled binary selection
variables and non-convex constraints. The reformulated problem facilitates the
design of an iterative resource allocation algorithm which obtains an optimal
solution based on the generalized Bender's decomposition (GBD) and serves as a
benchmark scheme. Furthermore, to strike a balance between computational
complexity and system performance, a suboptimal algorithm with polynomial time
complexity is proposed. Simulation results illustrate that the proposed GBD
based iterative algorithm converges to the global optimal solution and the
suboptimal algorithm achieves a close-to-optimal performance. Our results also
demonstrate the trade-off between power efficiency and the number of active
transmit antennas when the circuit power consumption is taken into account. In
particular, activating an exceedingly large number of antennas may not be a
power efficient solution for reducing the total system power consumption. In
addition, our results reveal that FD systems facilitate significant power
savings compared to traditional half-duplex systems, despite the non-negligible
self-interference.
|
1502.02171 | Liang Zheng | Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, Jiahao Bu, Qi Tian | Person Re-identification Meets Image Search | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For long time, person re-identification and image search are two separately
studied tasks. However, for person re-identification, the effectiveness of
local features and the "query-search" mode make it well posed for image search
techniques.
In the light of recent advances in image search, this paper proposes to treat
person re-identification as an image search problem. Specifically, this paper
claims two major contributions. 1) By designing an unsupervised Bag-of-Words
representation, we are devoted to bridging the gap between the two tasks by
integrating techniques from image search in person re-identification. We show
that our system sets up an effective yet efficient baseline that is amenable to
further supervised/unsupervised improvements. 2) We contribute a new high
quality dataset which uses DPM detector and includes a number of distractor
images. Our dataset reaches closer to realistic settings, and new perspectives
are provided.
Compared with approaches that rely on feature-feature match, our method is
faster by over two orders of magnitude. Moreover, on three datasets, we report
competitive results compared with the state-of-the-art methods.
| [
{
"version": "v1",
"created": "Sat, 7 Feb 2015 18:56:35 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Zheng",
"Liang",
""
],
[
"Shen",
"Liyue",
""
],
[
"Tian",
"Lu",
""
],
[
"Wang",
"Shengjin",
""
],
[
"Bu",
"Jiahao",
""
],
[
"Tian",
"Qi",
""
]
] | TITLE: Person Re-identification Meets Image Search
ABSTRACT: For long time, person re-identification and image search are two separately
studied tasks. However, for person re-identification, the effectiveness of
local features and the "query-search" mode make it well posed for image search
techniques.
In the light of recent advances in image search, this paper proposes to treat
person re-identification as an image search problem. Specifically, this paper
claims two major contributions. 1) By designing an unsupervised Bag-of-Words
representation, we are devoted to bridging the gap between the two tasks by
integrating techniques from image search in person re-identification. We show
that our system sets up an effective yet efficient baseline that is amenable to
further supervised/unsupervised improvements. 2) We contribute a new high
quality dataset which uses DPM detector and includes a number of distractor
images. Our dataset reaches closer to realistic settings, and new perspectives
are provided.
Compared with approaches that rely on feature-feature match, our method is
faster by over two orders of magnitude. Moreover, on three datasets, we report
competitive results compared with the state-of-the-art methods.
|
1502.02215 | Jobin Wilson | Jobin Wilson, Chitharanj Kachappilly, Rakesh Mohan, Prateek Kapadia,
Arun Soman, Santanu Chaudhury | Real World Applications of Machine Learning Techniques over Large Mobile
Subscriber Datasets | SE4ML: Software Engineering for Machine Learning (NIPS 2014 Workshop)
https://sites.google.com/site/software4ml/accepted-papers | null | null | null | cs.LG cs.CY cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Communication Service Providers (CSPs) are in a unique position to utilize
their vast transactional data assets generated from interactions of subscribers
with network elements as well as with other subscribers. CSPs could leverage
its data assets for a gamut of applications such as service personalization,
predictive offer management, loyalty management, revenue forecasting, network
capacity planning, product bundle optimization and churn management to gain
significant competitive advantage. However, due to the sheer data volume,
variety, velocity and veracity of mobile subscriber datasets, sophisticated
data analytics techniques and frameworks are necessary to derive actionable
insights in a useable timeframe. In this paper, we describe our journey from a
relational database management system (RDBMS) based campaign management
solution which allowed data scientists and marketers to use hand-written rules
for service personalization and targeted promotions to a distributed Big Data
Analytics platform, capable of performing large scale machine learning and data
mining to deliver real time service personalization, predictive modelling and
product optimization. Our work involves a careful blend of technology,
processes and best practices, which facilitate man-machine collaboration and
continuous experimentation to derive measurable economic value from data. Our
platform has a reach of more than 500 million mobile subscribers worldwide,
delivering over 1 billion personalized recommendations annually, processing a
total data volume of 64 Petabytes, corresponding to 8.5 trillion events.
| [
{
"version": "v1",
"created": "Sun, 8 Feb 2015 06:18:55 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Wilson",
"Jobin",
""
],
[
"Kachappilly",
"Chitharanj",
""
],
[
"Mohan",
"Rakesh",
""
],
[
"Kapadia",
"Prateek",
""
],
[
"Soman",
"Arun",
""
],
[
"Chaudhury",
"Santanu",
""
]
] | TITLE: Real World Applications of Machine Learning Techniques over Large Mobile
Subscriber Datasets
ABSTRACT: Communication Service Providers (CSPs) are in a unique position to utilize
their vast transactional data assets generated from interactions of subscribers
with network elements as well as with other subscribers. CSPs could leverage
its data assets for a gamut of applications such as service personalization,
predictive offer management, loyalty management, revenue forecasting, network
capacity planning, product bundle optimization and churn management to gain
significant competitive advantage. However, due to the sheer data volume,
variety, velocity and veracity of mobile subscriber datasets, sophisticated
data analytics techniques and frameworks are necessary to derive actionable
insights in a useable timeframe. In this paper, we describe our journey from a
relational database management system (RDBMS) based campaign management
solution which allowed data scientists and marketers to use hand-written rules
for service personalization and targeted promotions to a distributed Big Data
Analytics platform, capable of performing large scale machine learning and data
mining to deliver real time service personalization, predictive modelling and
product optimization. Our work involves a careful blend of technology,
processes and best practices, which facilitate man-machine collaboration and
continuous experimentation to derive measurable economic value from data. Our
platform has a reach of more than 500 million mobile subscribers worldwide,
delivering over 1 billion personalized recommendations annually, processing a
total data volume of 64 Petabytes, corresponding to 8.5 trillion events.
|
1502.02407 | James J.Q. Yu | James J.Q. Yu and Victor O.K. Li | A Social Spider Algorithm for Global Optimization | null | null | null | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The growing complexity of real-world problems has motivated computer
scientists to search for efficient problem-solving methods. Metaheuristics
based on evolutionary computation and swarm intelligence are outstanding
examples of nature-inspired solution techniques. Inspired by the social
spiders, we propose a novel Social Spider Algorithm to solve global
optimization problems. This algorithm is mainly based on the foraging strategy
of social spiders, utilizing the vibrations on the spider web to determine the
positions of preys. Different from the previously proposed swarm intelligence
algorithms, we introduce a new social animal foraging strategy model to solve
optimization problems. In addition, we perform preliminary parameter
sensitivity analysis for our proposed algorithm, developing guidelines for
choosing the parameter values. The Social Spider Algorithm is evaluated by a
series of widely-used benchmark functions, and our proposed algorithm has
superior performance compared with other state-of-the-art metaheuristics.
| [
{
"version": "v1",
"created": "Mon, 9 Feb 2015 09:46:40 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Yu",
"James J. Q.",
""
],
[
"Li",
"Victor O. K.",
""
]
] | TITLE: A Social Spider Algorithm for Global Optimization
ABSTRACT: The growing complexity of real-world problems has motivated computer
scientists to search for efficient problem-solving methods. Metaheuristics
based on evolutionary computation and swarm intelligence are outstanding
examples of nature-inspired solution techniques. Inspired by the social
spiders, we propose a novel Social Spider Algorithm to solve global
optimization problems. This algorithm is mainly based on the foraging strategy
of social spiders, utilizing the vibrations on the spider web to determine the
positions of preys. Different from the previously proposed swarm intelligence
algorithms, we introduce a new social animal foraging strategy model to solve
optimization problems. In addition, we perform preliminary parameter
sensitivity analysis for our proposed algorithm, developing guidelines for
choosing the parameter values. The Social Spider Algorithm is evaluated by a
series of widely-used benchmark functions, and our proposed algorithm has
superior performance compared with other state-of-the-art metaheuristics.
|
1502.02448 | Tom Crick | Tom Crick and Benjamin A. Hall and Samin Ishtiaq | Dear CAV, We Need to Talk About Reproducibility | Submitted to the 27th International Conference on Computer Aided
Verification (CAV 2015); 9 pages, LaTeX | null | null | null | cs.LO cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | How many times have you tried to re-implement a past CAV tool paper, and
failed?
Reliably reproducing published scientific discoveries has been acknowledged
as a barrier to scientific progress for some time but there remains only a
small subset of software available to support the specific needs of the
research community (i.e. beyond generic tools such as source code
repositories). In this paper we propose an infrastructure for enabling
reproducibility in our community, by automating the build, unit testing and
benchmarking of research software.
| [
{
"version": "v1",
"created": "Mon, 9 Feb 2015 11:59:40 GMT"
}
] | 2015-02-10T00:00:00 | [
[
"Crick",
"Tom",
""
],
[
"Hall",
"Benjamin A.",
""
],
[
"Ishtiaq",
"Samin",
""
]
] | TITLE: Dear CAV, We Need to Talk About Reproducibility
ABSTRACT: How many times have you tried to re-implement a past CAV tool paper, and
failed?
Reliably reproducing published scientific discoveries has been acknowledged
as a barrier to scientific progress for some time but there remains only a
small subset of software available to support the specific needs of the
research community (i.e. beyond generic tools such as source code
repositories). In this paper we propose an infrastructure for enabling
reproducibility in our community, by automating the build, unit testing and
benchmarking of research software.
|
1406.0167 | Saurabh Paul | Saurabh Paul, Malik Magdon-Ismail and Petros Drineas | Feature Selection for Linear SVM with Provable Guarantees | Appearing in Proceedings of 18th AISTATS, JMLR W&CP, vol 38, 2015 | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We give two provably accurate feature-selection techniques for the linear
SVM. The algorithms run in deterministic and randomized time respectively. Our
algorithms can be used in an unsupervised or supervised setting. The supervised
approach is based on sampling features from support vectors. We prove that the
margin in the feature space is preserved to within $\epsilon$-relative error of
the margin in the full feature space in the worst-case. In the unsupervised
setting, we also provide worst-case guarantees of the radius of the minimum
enclosing ball, thereby ensuring comparable generalization as in the full
feature space and resolving an open problem posed in Dasgupta et al. We present
extensive experiments on real-world datasets to support our theory and to
demonstrate that our method is competitive and often better than prior
state-of-the-art, for which there are no known provable guarantees.
| [
{
"version": "v1",
"created": "Sun, 1 Jun 2014 14:37:54 GMT"
},
{
"version": "v2",
"created": "Mon, 20 Oct 2014 14:20:00 GMT"
},
{
"version": "v3",
"created": "Fri, 6 Feb 2015 13:43:54 GMT"
}
] | 2015-02-09T00:00:00 | [
[
"Paul",
"Saurabh",
""
],
[
"Magdon-Ismail",
"Malik",
""
],
[
"Drineas",
"Petros",
""
]
] | TITLE: Feature Selection for Linear SVM with Provable Guarantees
ABSTRACT: We give two provably accurate feature-selection techniques for the linear
SVM. The algorithms run in deterministic and randomized time respectively. Our
algorithms can be used in an unsupervised or supervised setting. The supervised
approach is based on sampling features from support vectors. We prove that the
margin in the feature space is preserved to within $\epsilon$-relative error of
the margin in the full feature space in the worst-case. In the unsupervised
setting, we also provide worst-case guarantees of the radius of the minimum
enclosing ball, thereby ensuring comparable generalization as in the full
feature space and resolving an open problem posed in Dasgupta et al. We present
extensive experiments on real-world datasets to support our theory and to
demonstrate that our method is competitive and often better than prior
state-of-the-art, for which there are no known provable guarantees.
|
1409.5181 | Zhilin Zhang | Zhilin Zhang, Zhouyue Pi, Benyuan Liu | TROIKA: A General Framework for Heart Rate Monitoring Using Wrist-Type
Photoplethysmographic Signals During Intensive Physical Exercise | Matlab codes and data are available at:
https://sites.google.com/site/researchbyzhang/ | IEEE Transactions on Biomedical Engineering, vol. 62, no. 2, pp.
522-531, February 2015 | 10.1109/TBME.2014.2359372 | null | cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Heart rate monitoring using wrist-type photoplethysmographic (PPG) signals
during subjects' intensive exercise is a difficult problem, since the signals
are contaminated by extremely strong motion artifacts caused by subjects' hand
movements. So far few works have studied this problem. In this work, a general
framework, termed TROIKA, is proposed, which consists of signal decomposiTion
for denoising, sparse signal RecOnstructIon for high-resolution spectrum
estimation, and spectral peaK trAcking with verification. The TROIKA framework
has high estimation accuracy and is robust to strong motion artifacts. Many
variants can be straightforwardly derived from this framework. Experimental
results on datasets recorded from 12 subjects during fast running at the peak
speed of 15 km/hour showed that the average absolute error of heart rate
estimation was 2.34 beat per minute (BPM), and the Pearson correlation between
the estimates and the ground-truth of heart rate was 0.992. This framework is
of great values to wearable devices such as smart-watches which use PPG signals
to monitor heart rate for fitness.
| [
{
"version": "v1",
"created": "Thu, 18 Sep 2014 03:24:49 GMT"
},
{
"version": "v2",
"created": "Tue, 20 Jan 2015 07:47:00 GMT"
},
{
"version": "v3",
"created": "Fri, 6 Feb 2015 06:16:09 GMT"
}
] | 2015-02-09T00:00:00 | [
[
"Zhang",
"Zhilin",
""
],
[
"Pi",
"Zhouyue",
""
],
[
"Liu",
"Benyuan",
""
]
] | TITLE: TROIKA: A General Framework for Heart Rate Monitoring Using Wrist-Type
Photoplethysmographic Signals During Intensive Physical Exercise
ABSTRACT: Heart rate monitoring using wrist-type photoplethysmographic (PPG) signals
during subjects' intensive exercise is a difficult problem, since the signals
are contaminated by extremely strong motion artifacts caused by subjects' hand
movements. So far few works have studied this problem. In this work, a general
framework, termed TROIKA, is proposed, which consists of signal decomposiTion
for denoising, sparse signal RecOnstructIon for high-resolution spectrum
estimation, and spectral peaK trAcking with verification. The TROIKA framework
has high estimation accuracy and is robust to strong motion artifacts. Many
variants can be straightforwardly derived from this framework. Experimental
results on datasets recorded from 12 subjects during fast running at the peak
speed of 15 km/hour showed that the average absolute error of heart rate
estimation was 2.34 beat per minute (BPM), and the Pearson correlation between
the estimates and the ground-truth of heart rate was 0.992. This framework is
of great values to wearable devices such as smart-watches which use PPG signals
to monitor heart rate for fitness.
|
1412.1145 | Victor Pan | Victor Y. Pan | Matrix Multiplication, Trilinear Decompositions, APA Algorithms, and
Summation | 16 pages, 2 figures | null | null | null | cs.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Matrix multiplication (hereafter we use the acronym MM) is among the most
fundamental operations of modern computations. The efficiency of its
performance depends on various factors, in particular vectorization, data
movement and arithmetic complexity of the computations, but here we focus just
on the study of the arithmetic cost and the impact of this study on other areas
of modern computing. In the early 1970s it was expected that the
straightforward cubic time algorithm for MM will soon be accelerated to enable
MM in nearly quadratic arithmetic time, with some far fetched implications.
While pursuing this goal the mainstream research had its focus on the decrease
of the classical exponent 3 of the complexity of MM towards its lower bound 2,
disregarding the growth of the input size required to support this decrease.
Eventually, surprising combinations of novel ideas and sophisticated techniques
enabled the decrease of the exponent to its benchmark value of about 2.38, but
the supporting MM algorithms improved the straightforward one only for the
inputs of immense sizes. Meanwhile, the communication complexity, rather than
the arithmetic complexity, has become the bottleneck of computations in linear
algebra. This development may seem to undermine the value of the past and
future research aimed at the decrease of the arithmetic cost of MM, but we feel
that the study should be reassessed rather than closed and forgotten. We review
the old and new work in this area in the present day context, recall some major
techniques introduced in the study of MM, discuss their impact on the modern
theory and practice of computations for MM and beyond MM, and link one of these
techniques to some simple algorithms for inner product and summation.
| [
{
"version": "v1",
"created": "Wed, 3 Dec 2014 01:00:19 GMT"
},
{
"version": "v2",
"created": "Thu, 18 Dec 2014 14:07:37 GMT"
},
{
"version": "v3",
"created": "Thu, 5 Feb 2015 22:26:34 GMT"
}
] | 2015-02-09T00:00:00 | [
[
"Pan",
"Victor Y.",
""
]
] | TITLE: Matrix Multiplication, Trilinear Decompositions, APA Algorithms, and
Summation
ABSTRACT: Matrix multiplication (hereafter we use the acronym MM) is among the most
fundamental operations of modern computations. The efficiency of its
performance depends on various factors, in particular vectorization, data
movement and arithmetic complexity of the computations, but here we focus just
on the study of the arithmetic cost and the impact of this study on other areas
of modern computing. In the early 1970s it was expected that the
straightforward cubic time algorithm for MM will soon be accelerated to enable
MM in nearly quadratic arithmetic time, with some far fetched implications.
While pursuing this goal the mainstream research had its focus on the decrease
of the classical exponent 3 of the complexity of MM towards its lower bound 2,
disregarding the growth of the input size required to support this decrease.
Eventually, surprising combinations of novel ideas and sophisticated techniques
enabled the decrease of the exponent to its benchmark value of about 2.38, but
the supporting MM algorithms improved the straightforward one only for the
inputs of immense sizes. Meanwhile, the communication complexity, rather than
the arithmetic complexity, has become the bottleneck of computations in linear
algebra. This development may seem to undermine the value of the past and
future research aimed at the decrease of the arithmetic cost of MM, but we feel
that the study should be reassessed rather than closed and forgotten. We review
the old and new work in this area in the present day context, recall some major
techniques introduced in the study of MM, discuss their impact on the modern
theory and practice of computations for MM and beyond MM, and link one of these
techniques to some simple algorithms for inner product and summation.
|
1502.01782 | Conrad Sanderson | Johanna Carvajal, Conrad Sanderson, Chris McCool, Brian C. Lovell | Multi-Action Recognition via Stochastic Modelling of Optical Flow and
Gradients | null | Workshop on Machine Learning for Sensory Data Analysis (MLSDA),
pp. 19-24, 2014 | 10.1145/2689746.2689748 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we propose a novel approach to multi-action recognition that
performs joint segmentation and classification. This approach models each
action using a Gaussian mixture using robust low-dimensional action features.
Segmentation is achieved by performing classification on overlapping temporal
windows, which are then merged to produce the final result. This approach is
considerably less complicated than previous methods which use dynamic
programming or computationally expensive hidden Markov models (HMMs). Initial
experiments on a stitched version of the KTH dataset show that the proposed
approach achieves an accuracy of 78.3%, outperforming a recent HMM-based
approach which obtained 71.2%.
| [
{
"version": "v1",
"created": "Fri, 6 Feb 2015 03:30:10 GMT"
}
] | 2015-02-09T00:00:00 | [
[
"Carvajal",
"Johanna",
""
],
[
"Sanderson",
"Conrad",
""
],
[
"McCool",
"Chris",
""
],
[
"Lovell",
"Brian C.",
""
]
] | TITLE: Multi-Action Recognition via Stochastic Modelling of Optical Flow and
Gradients
ABSTRACT: In this paper we propose a novel approach to multi-action recognition that
performs joint segmentation and classification. This approach models each
action using a Gaussian mixture using robust low-dimensional action features.
Segmentation is achieved by performing classification on overlapping temporal
windows, which are then merged to produce the final result. This approach is
considerably less complicated than previous methods which use dynamic
programming or computationally expensive hidden Markov models (HMMs). Initial
experiments on a stitched version of the KTH dataset show that the proposed
approach achieves an accuracy of 78.3%, outperforming a recent HMM-based
approach which obtained 71.2%.
|
1502.01801 | Sayan Mitra | Chuchu Fan and Sayan Mitra | Bounded Verification with On-the-Fly Discrepancy Computation | 24 pages | null | null | University of Illinois Urbana Champaign, Tech Report
UILU-ENG-15-2201 | cs.SY cs.NA math.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Simulation-based verification algorithms can provide formal safety guarantees
for nonlinear and hybrid systems. The previous algorithms rely on user provided
model annotations called discrepancy function, which are crucial for computing
reachtubes from simulations. In this paper, we eliminate this requirement by
presenting an algorithm for computing piece-wise exponential discrepancy
functions. The algorithm relies on computing local convergence or divergence
rates of trajectories along a simulation using a coarse over-approximation of
the reach set and bounding the maximal eigenvalue of the Jacobian over this
over-approximation. The resulting discrepancy function preserves the soundness
and the relative completeness of the verification algorithm. We also provide a
coordinate transformation method to improve the local estimates for the
convergence or divergence rates in practical examples. We extend the method to
get the input-to-state discrepancy of nonlinear dynamical systems which can be
used for compositional analysis. Our experiments show that the approach is
effective in terms of running time for several benchmark problems, scales
reasonably to larger dimensional systems, and compares favorably with respect
to available tools for nonlinear models.
| [
{
"version": "v1",
"created": "Fri, 6 Feb 2015 05:32:03 GMT"
}
] | 2015-02-09T00:00:00 | [
[
"Fan",
"Chuchu",
""
],
[
"Mitra",
"Sayan",
""
]
] | TITLE: Bounded Verification with On-the-Fly Discrepancy Computation
ABSTRACT: Simulation-based verification algorithms can provide formal safety guarantees
for nonlinear and hybrid systems. The previous algorithms rely on user provided
model annotations called discrepancy function, which are crucial for computing
reachtubes from simulations. In this paper, we eliminate this requirement by
presenting an algorithm for computing piece-wise exponential discrepancy
functions. The algorithm relies on computing local convergence or divergence
rates of trajectories along a simulation using a coarse over-approximation of
the reach set and bounding the maximal eigenvalue of the Jacobian over this
over-approximation. The resulting discrepancy function preserves the soundness
and the relative completeness of the verification algorithm. We also provide a
coordinate transformation method to improve the local estimates for the
convergence or divergence rates in practical examples. We extend the method to
get the input-to-state discrepancy of nonlinear dynamical systems which can be
used for compositional analysis. Our experiments show that the approach is
effective in terms of running time for several benchmark problems, scales
reasonably to larger dimensional systems, and compares favorably with respect
to available tools for nonlinear models.
|
1502.01812 | Teng Li Dr. | Teng Li, Huan Chang, Meng Wang, Bingbing Ni, Richang Hong and
Shuicheng Yan | Crowded Scene Analysis: A Survey | 20 pages in IEEE Transactions on Circuits and Systems for Video
Technology, 2015 | null | 10.1109/TCSVT.2014.2358029 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Automated scene analysis has been a topic of great interest in computer
vision and cognitive science. Recently, with the growth of crowd phenomena in
the real world, crowded scene analysis has attracted much attention. However,
the visual occlusions and ambiguities in crowded scenes, as well as the complex
behaviors and scene semantics, make the analysis a challenging task. In the
past few years, an increasing number of works on crowded scene analysis have
been reported, covering different aspects including crowd motion pattern
learning, crowd behavior and activity analysis, and anomaly detection in
crowds. This paper surveys the state-of-the-art techniques on this topic. We
first provide the background knowledge and the available features related to
crowded scenes. Then, existing models, popular algorithms, evaluation
protocols, as well as system performance are provided corresponding to
different aspects of crowded scene analysis. We also outline the available
datasets for performance evaluation. Finally, some research problems and
promising future directions are presented with discussions.
| [
{
"version": "v1",
"created": "Fri, 6 Feb 2015 06:36:12 GMT"
}
] | 2015-02-09T00:00:00 | [
[
"Li",
"Teng",
""
],
[
"Chang",
"Huan",
""
],
[
"Wang",
"Meng",
""
],
[
"Ni",
"Bingbing",
""
],
[
"Hong",
"Richang",
""
],
[
"Yan",
"Shuicheng",
""
]
] | TITLE: Crowded Scene Analysis: A Survey
ABSTRACT: Automated scene analysis has been a topic of great interest in computer
vision and cognitive science. Recently, with the growth of crowd phenomena in
the real world, crowded scene analysis has attracted much attention. However,
the visual occlusions and ambiguities in crowded scenes, as well as the complex
behaviors and scene semantics, make the analysis a challenging task. In the
past few years, an increasing number of works on crowded scene analysis have
been reported, covering different aspects including crowd motion pattern
learning, crowd behavior and activity analysis, and anomaly detection in
crowds. This paper surveys the state-of-the-art techniques on this topic. We
first provide the background knowledge and the available features related to
crowded scenes. Then, existing models, popular algorithms, evaluation
protocols, as well as system performance are provided corresponding to
different aspects of crowded scene analysis. We also outline the available
datasets for performance evaluation. Finally, some research problems and
promising future directions are presented with discussions.
|
1502.01827 | Guang-Tong Zhou | Guang-Tong Zhou, Sung Ju Hwang, Mark Schmidt, Leonid Sigal and Greg
Mori | Hierarchical Maximum-Margin Clustering | null | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a hierarchical maximum-margin clustering method for unsupervised
data analysis. Our method extends beyond flat maximum-margin clustering, and
performs clustering recursively in a top-down manner. We propose an effective
greedy splitting criteria for selecting which cluster to split next, and employ
regularizers that enforce feature sharing/competition for capturing data
semantics. Experimental results obtained on four standard datasets show that
our method outperforms flat and hierarchical clustering baselines, while
forming clean and semantically meaningful cluster hierarchies.
| [
{
"version": "v1",
"created": "Fri, 6 Feb 2015 08:37:55 GMT"
}
] | 2015-02-09T00:00:00 | [
[
"Zhou",
"Guang-Tong",
""
],
[
"Hwang",
"Sung Ju",
""
],
[
"Schmidt",
"Mark",
""
],
[
"Sigal",
"Leonid",
""
],
[
"Mori",
"Greg",
""
]
] | TITLE: Hierarchical Maximum-Margin Clustering
ABSTRACT: We present a hierarchical maximum-margin clustering method for unsupervised
data analysis. Our method extends beyond flat maximum-margin clustering, and
performs clustering recursively in a top-down manner. We propose an effective
greedy splitting criteria for selecting which cluster to split next, and employ
regularizers that enforce feature sharing/competition for capturing data
semantics. Experimental results obtained on four standard datasets show that
our method outperforms flat and hierarchical clustering baselines, while
forming clean and semantically meaningful cluster hierarchies.
|
1502.01852 | Kaiming He | Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun | Delving Deep into Rectifiers: Surpassing Human-Level Performance on
ImageNet Classification | null | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Rectified activation units (rectifiers) are essential for state-of-the-art
neural networks. In this work, we study rectifier neural networks for image
classification from two aspects. First, we propose a Parametric Rectified
Linear Unit (PReLU) that generalizes the traditional rectified unit. PReLU
improves model fitting with nearly zero extra computational cost and little
overfitting risk. Second, we derive a robust initialization method that
particularly considers the rectifier nonlinearities. This method enables us to
train extremely deep rectified models directly from scratch and to investigate
deeper or wider network architectures. Based on our PReLU networks
(PReLU-nets), we achieve 4.94% top-5 test error on the ImageNet 2012
classification dataset. This is a 26% relative improvement over the ILSVRC 2014
winner (GoogLeNet, 6.66%). To our knowledge, our result is the first to surpass
human-level performance (5.1%, Russakovsky et al.) on this visual recognition
challenge.
| [
{
"version": "v1",
"created": "Fri, 6 Feb 2015 10:44:00 GMT"
}
] | 2015-02-09T00:00:00 | [
[
"He",
"Kaiming",
""
],
[
"Zhang",
"Xiangyu",
""
],
[
"Ren",
"Shaoqing",
""
],
[
"Sun",
"Jian",
""
]
] | TITLE: Delving Deep into Rectifiers: Surpassing Human-Level Performance on
ImageNet Classification
ABSTRACT: Rectified activation units (rectifiers) are essential for state-of-the-art
neural networks. In this work, we study rectifier neural networks for image
classification from two aspects. First, we propose a Parametric Rectified
Linear Unit (PReLU) that generalizes the traditional rectified unit. PReLU
improves model fitting with nearly zero extra computational cost and little
overfitting risk. Second, we derive a robust initialization method that
particularly considers the rectifier nonlinearities. This method enables us to
train extremely deep rectified models directly from scratch and to investigate
deeper or wider network architectures. Based on our PReLU networks
(PReLU-nets), we achieve 4.94% top-5 test error on the ImageNet 2012
classification dataset. This is a 26% relative improvement over the ILSVRC 2014
winner (GoogLeNet, 6.66%). To our knowledge, our result is the first to surpass
human-level performance (5.1%, Russakovsky et al.) on this visual recognition
challenge.
|
1502.01972 | Michael Saint-Guillain | Michael Saint-Guillain, Yves Deville and Christine Solnon | A Multistage Stochastic Programming Approach to the Dynamic and
Stochastic VRPTW - Extended version | Extended version of the same-name study submitted for publication in
conference CPAIOR2015 | null | null | null | cs.AI cs.DM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider a dynamic vehicle routing problem with time windows and
stochastic customers (DS-VRPTW), such that customers may request for services
as vehicles have already started their tours. To solve this problem, the goal
is to provide a decision rule for choosing, at each time step, the next action
to perform in light of known requests and probabilistic knowledge on requests
likelihood. We introduce a new decision rule, called Global Stochastic
Assessment (GSA) rule for the DS-VRPTW, and we compare it with existing
decision rules, such as MSA. In particular, we show that GSA fully integrates
nonanticipativity constraints so that it leads to better decisions in our
stochastic context. We describe a new heuristic approach for efficiently
approximating our GSA rule. We introduce a new waiting strategy. Experiments on
dynamic and stochastic benchmarks, which include instances of different degrees
of dynamism, show that not only our approach is competitive with
state-of-the-art methods, but also enables to compute meaningful offline
solutions to fully dynamic problems where absolutely no a priori customer
request is provided.
| [
{
"version": "v1",
"created": "Fri, 6 Feb 2015 18:11:09 GMT"
}
] | 2015-02-09T00:00:00 | [
[
"Saint-Guillain",
"Michael",
""
],
[
"Deville",
"Yves",
""
],
[
"Solnon",
"Christine",
""
]
] | TITLE: A Multistage Stochastic Programming Approach to the Dynamic and
Stochastic VRPTW - Extended version
ABSTRACT: We consider a dynamic vehicle routing problem with time windows and
stochastic customers (DS-VRPTW), such that customers may request for services
as vehicles have already started their tours. To solve this problem, the goal
is to provide a decision rule for choosing, at each time step, the next action
to perform in light of known requests and probabilistic knowledge on requests
likelihood. We introduce a new decision rule, called Global Stochastic
Assessment (GSA) rule for the DS-VRPTW, and we compare it with existing
decision rules, such as MSA. In particular, we show that GSA fully integrates
nonanticipativity constraints so that it leads to better decisions in our
stochastic context. We describe a new heuristic approach for efficiently
approximating our GSA rule. We introduce a new waiting strategy. Experiments on
dynamic and stochastic benchmarks, which include instances of different degrees
of dynamism, show that not only our approach is competitive with
state-of-the-art methods, but also enables to compute meaningful offline
solutions to fully dynamic problems where absolutely no a priori customer
request is provided.
|
1403.6025 | Ruven Pillay | Emmanuel Bertin, Ruven Pillay, Chiara Marmo | Web-Based Visualization of Very Large Scientific Astronomy Imagery | Published in Astronomy & Computing. IIPImage server available from
http://iipimage.sourceforge.net . Visiomatic code and demos available from
http://www.visiomatic.org/ | Astronomy and Computing, vol. 10, pp. 43-53, Apr. 2015 | 10.1016/j.ascom.2014.12.006 | null | astro-ph.IM cs.CE cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visualizing and navigating through large astronomy images from a remote
location with current astronomy display tools can be a frustrating experience
in terms of speed and ergonomics, especially on mobile devices. In this paper,
we present a high performance, versatile and robust client-server system for
remote visualization and analysis of extremely large scientific images.
Applications of this work include survey image quality control, interactive
data query and exploration, citizen science, as well as public outreach. The
proposed software is entirely open source and is designed to be generic and
applicable to a variety of datasets. It provides access to floating point data
at terabyte scales, with the ability to precisely adjust image settings in
real-time. The proposed clients are light-weight, platform-independent web
applications built on standard HTML5 web technologies and compatible with both
touch and mouse-based devices. We put the system to the test and assess the
performance of the system and show that a single server can comfortably handle
more than a hundred simultaneous users accessing full precision 32 bit
astronomy data.
| [
{
"version": "v1",
"created": "Mon, 24 Mar 2014 16:24:57 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jan 2015 14:21:56 GMT"
},
{
"version": "v3",
"created": "Sun, 1 Feb 2015 22:29:40 GMT"
},
{
"version": "v4",
"created": "Thu, 5 Feb 2015 10:40:31 GMT"
}
] | 2015-02-06T00:00:00 | [
[
"Bertin",
"Emmanuel",
""
],
[
"Pillay",
"Ruven",
""
],
[
"Marmo",
"Chiara",
""
]
] | TITLE: Web-Based Visualization of Very Large Scientific Astronomy Imagery
ABSTRACT: Visualizing and navigating through large astronomy images from a remote
location with current astronomy display tools can be a frustrating experience
in terms of speed and ergonomics, especially on mobile devices. In this paper,
we present a high performance, versatile and robust client-server system for
remote visualization and analysis of extremely large scientific images.
Applications of this work include survey image quality control, interactive
data query and exploration, citizen science, as well as public outreach. The
proposed software is entirely open source and is designed to be generic and
applicable to a variety of datasets. It provides access to floating point data
at terabyte scales, with the ability to precisely adjust image settings in
real-time. The proposed clients are light-weight, platform-independent web
applications built on standard HTML5 web technologies and compatible with both
touch and mouse-based devices. We put the system to the test and assess the
performance of the system and show that a single server can comfortably handle
more than a hundred simultaneous users accessing full precision 32 bit
astronomy data.
|
1410.6973 | Anna Choromanska | Mariusz Bojarski, Anna Choromanska, Krzysztof Choromanski, Yann LeCun | Differentially- and non-differentially-private random decision trees | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider supervised learning with random decision trees, where the tree
construction is completely random. The method is popularly used and works well
in practice despite the simplicity of the setting, but its statistical
mechanism is not yet well-understood. In this paper we provide strong
theoretical guarantees regarding learning with random decision trees. We
analyze and compare three different variants of the algorithm that have minimal
memory requirements: majority voting, threshold averaging and probabilistic
averaging. The random structure of the tree enables us to adapt these methods
to a differentially-private setting thus we also propose differentially-private
versions of all three schemes. We give upper-bounds on the generalization error
and mathematically explain how the accuracy depends on the number of random
decision trees. Furthermore, we prove that only logarithmic (in the size of the
dataset) number of independently selected random decision trees suffice to
correctly classify most of the data, even when differential-privacy guarantees
must be maintained. We empirically show that majority voting and threshold
averaging give the best accuracy, also for conservative users requiring high
privacy guarantees. Furthermore, we demonstrate that a simple majority voting
rule is an especially good candidate for the differentially-private classifier
since it is much less sensitive to the choice of forest parameters than other
methods.
| [
{
"version": "v1",
"created": "Sun, 26 Oct 2014 00:16:16 GMT"
},
{
"version": "v2",
"created": "Thu, 5 Feb 2015 20:48:11 GMT"
}
] | 2015-02-06T00:00:00 | [
[
"Bojarski",
"Mariusz",
""
],
[
"Choromanska",
"Anna",
""
],
[
"Choromanski",
"Krzysztof",
""
],
[
"LeCun",
"Yann",
""
]
] | TITLE: Differentially- and non-differentially-private random decision trees
ABSTRACT: We consider supervised learning with random decision trees, where the tree
construction is completely random. The method is popularly used and works well
in practice despite the simplicity of the setting, but its statistical
mechanism is not yet well-understood. In this paper we provide strong
theoretical guarantees regarding learning with random decision trees. We
analyze and compare three different variants of the algorithm that have minimal
memory requirements: majority voting, threshold averaging and probabilistic
averaging. The random structure of the tree enables us to adapt these methods
to a differentially-private setting thus we also propose differentially-private
versions of all three schemes. We give upper-bounds on the generalization error
and mathematically explain how the accuracy depends on the number of random
decision trees. Furthermore, we prove that only logarithmic (in the size of the
dataset) number of independently selected random decision trees suffice to
correctly classify most of the data, even when differential-privacy guarantees
must be maintained. We empirically show that majority voting and threshold
averaging give the best accuracy, also for conservative users requiring high
privacy guarantees. Furthermore, we demonstrate that a simple majority voting
rule is an especially good candidate for the differentially-private classifier
since it is much less sensitive to the choice of forest parameters than other
methods.
|
1311.1266 | Diego Amancio | Diego R. Amancio and Osvaldo N. Oliveira Jr. and Luciano da F. Costa | Topological-collaborative approach for disambiguating authors' names in
collaborative networks | To appear in Scientometrics, 2014 | Scientometrics 102 (1), 465--485, 2015 | 10.1007/s11192-014-1381-9 | null | cs.SI physics.soc-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Concepts and methods of complex networks have been employed to uncover
patterns in a myriad of complex systems. Unfortunately, the relevance and
significance of these patterns strongly depends on the reliability of the data
sets. In the study of collaboration networks, for instance, unavoidable noise
pervading author's collaboration datasets arises when authors share the same
name. To address this problem, we derive a hybrid approach based on authors'
collaboration patterns and on topological features of collaborative networks.
Our results show that the combination of strategies, in most cases, performs
better than the traditional approach which disregards topological features. We
also show that the main factor for improving the discriminability of homonymous
authors is the average distance between authors. Finally, we show that it is
possible to predict the weighting associated to each strategy compounding the
hybrid system by examining the discrimination obtained from the traditional
analysis of collaboration patterns. Once the methodology devised here is
generic, our approach is potentially useful to classify many other networked
systems governed by complex interactions.
| [
{
"version": "v1",
"created": "Wed, 6 Nov 2013 01:43:51 GMT"
},
{
"version": "v2",
"created": "Mon, 30 Jun 2014 16:01:45 GMT"
}
] | 2015-02-05T00:00:00 | [
[
"Amancio",
"Diego R.",
""
],
[
"Oliveira",
"Osvaldo N.",
"Jr."
],
[
"Costa",
"Luciano da F.",
""
]
] | TITLE: Topological-collaborative approach for disambiguating authors' names in
collaborative networks
ABSTRACT: Concepts and methods of complex networks have been employed to uncover
patterns in a myriad of complex systems. Unfortunately, the relevance and
significance of these patterns strongly depends on the reliability of the data
sets. In the study of collaboration networks, for instance, unavoidable noise
pervading author's collaboration datasets arises when authors share the same
name. To address this problem, we derive a hybrid approach based on authors'
collaboration patterns and on topological features of collaborative networks.
Our results show that the combination of strategies, in most cases, performs
better than the traditional approach which disregards topological features. We
also show that the main factor for improving the discriminability of homonymous
authors is the average distance between authors. Finally, we show that it is
possible to predict the weighting associated to each strategy compounding the
hybrid system by examining the discrimination obtained from the traditional
analysis of collaboration patterns. Once the methodology devised here is
generic, our approach is potentially useful to classify many other networked
systems governed by complex interactions.
|
1403.2407 | Lorenzo Reyes | Andrey Bernstein (1), Lorenzo Reyes-Chamorro (2), Jean-Yves Le Boudec
(1) and Mario Paolone (2) ((1) Laboratory for Communications and Applications
2. \'Ecole Polytechnique F\'ed\'erale de Lausanne, (2) Distributed Electrical
Systems Laboratory. \'Ecole Polytechnique F\'ed\'erale de Lausanne) | A Composable Method for Real-Time Control of Active Distribution
Networks with Explicit Power Setpoints | 71 pages, 21 figures, 5 tables | null | null | null | cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The conventional approach for the control of distribution networks, in the
presence of active generation and/or controllable loads and storage, involves a
combination of both frequency and voltage regulation at different time scales.
With the increased penetration of stochastic resources, distributed generation
and demand response, this approach shows severe limitations in both the optimal
and feasible operation of these networks, as well as in the aggregation of the
network resources for upper-layer power systems. An alternative approach is to
directly control the targeted grid by defining explicit and real-time setpoints
for active/reactive power absorptions/injections defined by a solution of a
specific optimization problem; but this quickly becomes intractable when
systems get large or diverse. In this paper, we address this problem and
propose a method for the explicit control of the grid status, based on a common
abstract model characterized by the main property of being composable. That is
to say, subsystems can be aggregated into virtual devices that hide their
internal complexity. Thus the proposed method can easily cope with systems of
any size or complexity. The framework is presented in this Part I, whilst in
Part II we illustrate its application to a CIGR\'E low voltage benchmark
microgrid. In particular, we provide implementation examples with respect to
typical devices connected to distribution networks and evaluate of the
performance and benefits of the proposed control framework.
| [
{
"version": "v1",
"created": "Fri, 7 Mar 2014 13:08:54 GMT"
},
{
"version": "v2",
"created": "Tue, 6 May 2014 15:16:04 GMT"
},
{
"version": "v3",
"created": "Mon, 11 Aug 2014 17:26:26 GMT"
},
{
"version": "v4",
"created": "Wed, 4 Feb 2015 00:09:54 GMT"
}
] | 2015-02-05T00:00:00 | [
[
"Bernstein",
"Andrey",
""
],
[
"Reyes-Chamorro",
"Lorenzo",
""
],
[
"Boudec",
"Jean-Yves Le",
""
],
[
"Paolone",
"Mario",
""
]
] | TITLE: A Composable Method for Real-Time Control of Active Distribution
Networks with Explicit Power Setpoints
ABSTRACT: The conventional approach for the control of distribution networks, in the
presence of active generation and/or controllable loads and storage, involves a
combination of both frequency and voltage regulation at different time scales.
With the increased penetration of stochastic resources, distributed generation
and demand response, this approach shows severe limitations in both the optimal
and feasible operation of these networks, as well as in the aggregation of the
network resources for upper-layer power systems. An alternative approach is to
directly control the targeted grid by defining explicit and real-time setpoints
for active/reactive power absorptions/injections defined by a solution of a
specific optimization problem; but this quickly becomes intractable when
systems get large or diverse. In this paper, we address this problem and
propose a method for the explicit control of the grid status, based on a common
abstract model characterized by the main property of being composable. That is
to say, subsystems can be aggregated into virtual devices that hide their
internal complexity. Thus the proposed method can easily cope with systems of
any size or complexity. The framework is presented in this Part I, whilst in
Part II we illustrate its application to a CIGR\'E low voltage benchmark
microgrid. In particular, we provide implementation examples with respect to
typical devices connected to distribution networks and evaluate of the
performance and benefits of the proposed control framework.
|
1501.03396 | Davide Fransos | Luca Bruno and Davide Fransos | Sand transverse dune aerodynamics: 3D Coherent Flow Structures from a
computational study | null | null | null | null | physics.flu-dyn cs.CE physics.ao-ph physics.geo-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The engineering interest about dune fields is dictated by the their
interaction with a number of human infrastructures in arid environments. Sand
dunes dynamics is dictated by wind and its ability to induce sand erosion,
transport and deposition. A deep understanding of dune aerodynamics serves then
to ground effective strategies for the protection of human infrastructures from
sand, the so-called sand mitigation. Because of their simple geometry and their
frequent occurrence in desert area, transverse sand dunes are usually adopted
in literature as a benchmark to investigate dune aerodynamics by means of both
computational or experimental approaches, usually in nominally 2D setups. The
present study aims at evaluating 3D flow features in the wake of a idealised
transverse dune, if any, under different nominally 2D setup conditions by means
of computational simulations and to compare the obtained results with
experimental measurements available in literature.
| [
{
"version": "v1",
"created": "Sun, 11 Jan 2015 13:31:40 GMT"
},
{
"version": "v2",
"created": "Wed, 4 Feb 2015 12:03:15 GMT"
}
] | 2015-02-05T00:00:00 | [
[
"Bruno",
"Luca",
""
],
[
"Fransos",
"Davide",
""
]
] | TITLE: Sand transverse dune aerodynamics: 3D Coherent Flow Structures from a
computational study
ABSTRACT: The engineering interest about dune fields is dictated by the their
interaction with a number of human infrastructures in arid environments. Sand
dunes dynamics is dictated by wind and its ability to induce sand erosion,
transport and deposition. A deep understanding of dune aerodynamics serves then
to ground effective strategies for the protection of human infrastructures from
sand, the so-called sand mitigation. Because of their simple geometry and their
frequent occurrence in desert area, transverse sand dunes are usually adopted
in literature as a benchmark to investigate dune aerodynamics by means of both
computational or experimental approaches, usually in nominally 2D setups. The
present study aims at evaluating 3D flow features in the wake of a idealised
transverse dune, if any, under different nominally 2D setup conditions by means
of computational simulations and to compare the obtained results with
experimental measurements available in literature.
|
1502.00254 | Ravi Kiran Sarvadevabhatla | Ravi Kiran Sarvadevabhatla and R. Venkatesh Babu | Freehand Sketch Recognition Using Deep Features | Submitted to ICIP-2015, 5 pages. Removed an erroneous claim regarding
a work cited in the paper | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Freehand sketches often contain sparse visual detail. In spite of the
sparsity, they are easily and consistently recognized by humans across
cultures, languages and age groups. Therefore, analyzing such sparse sketches
can aid our understanding of the neuro-cognitive processes involved in visual
representation and recognition. In the recent past, Convolutional Neural
Networks (CNNs) have emerged as a powerful framework for feature representation
and recognition for a variety of image domains. However, the domain of sketch
images has not been explored. This paper introduces a freehand sketch
recognition framework based on "deep" features extracted from CNNs. We use two
popular CNNs for our experiments -- Imagenet CNN and a modified version of
LeNet CNN. We evaluate our recognition framework on a publicly available
benchmark database containing thousands of freehand sketches depicting everyday
objects. Our results are an improvement over the existing state-of-the-art
accuracies by 3% - 11%. The effectiveness and relative compactness of our deep
features also make them an ideal candidate for related problems such as
sketch-based image retrieval. In addition, we provide a preliminary glimpse of
how such features can help identify crucial attributes (e.g. object-parts) of
the sketched objects.
| [
{
"version": "v1",
"created": "Sun, 1 Feb 2015 13:35:44 GMT"
},
{
"version": "v2",
"created": "Wed, 4 Feb 2015 14:41:15 GMT"
}
] | 2015-02-05T00:00:00 | [
[
"Sarvadevabhatla",
"Ravi Kiran",
""
],
[
"Babu",
"R. Venkatesh",
""
]
] | TITLE: Freehand Sketch Recognition Using Deep Features
ABSTRACT: Freehand sketches often contain sparse visual detail. In spite of the
sparsity, they are easily and consistently recognized by humans across
cultures, languages and age groups. Therefore, analyzing such sparse sketches
can aid our understanding of the neuro-cognitive processes involved in visual
representation and recognition. In the recent past, Convolutional Neural
Networks (CNNs) have emerged as a powerful framework for feature representation
and recognition for a variety of image domains. However, the domain of sketch
images has not been explored. This paper introduces a freehand sketch
recognition framework based on "deep" features extracted from CNNs. We use two
popular CNNs for our experiments -- Imagenet CNN and a modified version of
LeNet CNN. We evaluate our recognition framework on a publicly available
benchmark database containing thousands of freehand sketches depicting everyday
objects. Our results are an improvement over the existing state-of-the-art
accuracies by 3% - 11%. The effectiveness and relative compactness of our deep
features also make them an ideal candidate for related problems such as
sketch-based image retrieval. In addition, we provide a preliminary glimpse of
how such features can help identify crucial attributes (e.g. object-parts) of
the sketched objects.
|
1305.0194 | Chantal Cherifi | Cihan Aksoy, Vincent Labatut, Chantal Cherifi, Jean-Fran\c{c}ois
Santucci | MATAWS: A Multimodal Approach for Automatic WS Semantic Annotation | null | In International Conference on Networked Digital Technologies,
Springer CCIS 136),China (2011) | 10.1007/978-3-642-22185-9_27 | null | cs.SE cs.CL cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Many recent works aim at developing methods and tools for the processing of
semantic Web services. In order to be properly tested, these tools must be
applied to an appropriate benchmark, taking the form of a collection of
semantic WS descriptions. However, all of the existing publicly available
collections are limited by their size or their realism (use of randomly
generated or resampled descriptions). Larger and realistic syntactic (WSDL)
collections exist, but their semantic annotation requires a certain level of
automation, due to the number of operations to be processed. In this article,
we propose a fully automatic method to semantically annotate such large WS
collections. Our approach is multimodal, in the sense it takes advantage of the
latent semantics present not only in the parameter names, but also in the type
names and structures. Concept-to-word association is performed by using Sigma,
a mapping of WordNet to the SUMO ontology. After having described in details
our annotation method, we apply it to the larger collection of real-world
syntactic WS descriptions we could find, and assess its efficiency.
| [
{
"version": "v1",
"created": "Wed, 1 May 2013 15:07:18 GMT"
}
] | 2015-02-04T00:00:00 | [
[
"Aksoy",
"Cihan",
""
],
[
"Labatut",
"Vincent",
""
],
[
"Cherifi",
"Chantal",
""
],
[
"Santucci",
"Jean-François",
""
]
] | TITLE: MATAWS: A Multimodal Approach for Automatic WS Semantic Annotation
ABSTRACT: Many recent works aim at developing methods and tools for the processing of
semantic Web services. In order to be properly tested, these tools must be
applied to an appropriate benchmark, taking the form of a collection of
semantic WS descriptions. However, all of the existing publicly available
collections are limited by their size or their realism (use of randomly
generated or resampled descriptions). Larger and realistic syntactic (WSDL)
collections exist, but their semantic annotation requires a certain level of
automation, due to the number of operations to be processed. In this article,
we propose a fully automatic method to semantically annotate such large WS
collections. Our approach is multimodal, in the sense it takes advantage of the
latent semantics present not only in the parameter names, but also in the type
names and structures. Concept-to-word association is performed by using Sigma,
a mapping of WordNet to the SUMO ontology. After having described in details
our annotation method, we apply it to the larger collection of real-world
syntactic WS descriptions we could find, and assess its efficiency.
|
1502.00652 | Lubor Ladicky | \v{L}ubor Ladick\'y, Christian H\"ane and Marc Pollefeys | Learning the Matching Function | rejected from ACCV 2014 and probably from CVPR 2015 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The matching function for the problem of stereo reconstruction or optical
flow has been traditionally designed as a function of the distance between the
features describing matched pixels. This approach works under assumption, that
the appearance of pixels in two stereo cameras or in two consecutive video
frames does not change dramatically. However, this might not be the case, if we
try to match pixels over a large interval of time.
In this paper we propose a method, which learns the matching function, that
automatically finds the space of allowed changes in visual appearance, such as
due to the motion blur, chromatic distortions, different colour calibration or
seasonal changes. Furthermore, it automatically learns the importance of
matching scores of contextual features at different relative locations and
scales. Proposed classifier gives reliable estimations of pixel disparities
already without any form of regularization.
We evaluated our method on two standard problems - stereo matching on KITTI
outdoor dataset, optical flow on Sintel data set, and on newly introduced
TimeLapse change detection dataset. Our algorithm obtained very promising
results comparable to the state-of-the-art.
| [
{
"version": "v1",
"created": "Mon, 2 Feb 2015 21:20:43 GMT"
}
] | 2015-02-04T00:00:00 | [
[
"Ladický",
"Ľubor",
""
],
[
"Häne",
"Christian",
""
],
[
"Pollefeys",
"Marc",
""
]
] | TITLE: Learning the Matching Function
ABSTRACT: The matching function for the problem of stereo reconstruction or optical
flow has been traditionally designed as a function of the distance between the
features describing matched pixels. This approach works under assumption, that
the appearance of pixels in two stereo cameras or in two consecutive video
frames does not change dramatically. However, this might not be the case, if we
try to match pixels over a large interval of time.
In this paper we propose a method, which learns the matching function, that
automatically finds the space of allowed changes in visual appearance, such as
due to the motion blur, chromatic distortions, different colour calibration or
seasonal changes. Furthermore, it automatically learns the importance of
matching scores of contextual features at different relative locations and
scales. Proposed classifier gives reliable estimations of pixel disparities
already without any form of regularization.
We evaluated our method on two standard problems - stereo matching on KITTI
outdoor dataset, optical flow on Sintel data set, and on newly introduced
TimeLapse change detection dataset. Our algorithm obtained very promising
results comparable to the state-of-the-art.
|
1502.00712 | Liang Lin | Zhanglin Peng, Liang Lin, Ruimao Zhang, Jing Xu | Deep Boosting: Layered Feature Mining for General Image Classification | 6 pages, 4 figures, ICME 2014 | Multimedia and Expo (ICME), 2014 IEEE International Conference on
, vol., no., pp.1,6, 14-18 July 2014 | 10.1109/ICME.2014.6890323 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Constructing effective representations is a critical but challenging problem
in multimedia understanding. The traditional handcraft features often rely on
domain knowledge, limiting the performances of exiting methods. This paper
discusses a novel computational architecture for general image feature mining,
which assembles the primitive filters (i.e. Gabor wavelets) into compositional
features in a layer-wise manner. In each layer, we produce a number of base
classifiers (i.e. regression stumps) associated with the generated features,
and discover informative compositions by using the boosting algorithm. The
output compositional features of each layer are treated as the base components
to build up the next layer. Our framework is able to generate expressive image
representations while inducing very discriminate functions for image
classification. The experiments are conducted on several public datasets, and
we demonstrate superior performances over state-of-the-art approaches.
| [
{
"version": "v1",
"created": "Tue, 3 Feb 2015 02:44:10 GMT"
}
] | 2015-02-04T00:00:00 | [
[
"Peng",
"Zhanglin",
""
],
[
"Lin",
"Liang",
""
],
[
"Zhang",
"Ruimao",
""
],
[
"Xu",
"Jing",
""
]
] | TITLE: Deep Boosting: Layered Feature Mining for General Image Classification
ABSTRACT: Constructing effective representations is a critical but challenging problem
in multimedia understanding. The traditional handcraft features often rely on
domain knowledge, limiting the performances of exiting methods. This paper
discusses a novel computational architecture for general image feature mining,
which assembles the primitive filters (i.e. Gabor wavelets) into compositional
features in a layer-wise manner. In each layer, we produce a number of base
classifiers (i.e. regression stumps) associated with the generated features,
and discover informative compositions by using the boosting algorithm. The
output compositional features of each layer are treated as the base components
to build up the next layer. Our framework is able to generate expressive image
representations while inducing very discriminate functions for image
classification. The experiments are conducted on several public datasets, and
we demonstrate superior performances over state-of-the-art approaches.
|
1502.00717 | Hongyuan Zhu | Hongyuan Zhu, Fanman Meng, Jianfei Cai, Shijian Lu | Beyond Pixels: A Comprehensive Survey from Bottom-up to Semantic Image
Segmentation and Cosegmentation | submitted to Elsevier Journal of Visual Communications and Image
Representation | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/3.0/ | Image segmentation refers to the process to divide an image into
nonoverlapping meaningful regions according to human perception, which has
become a classic topic since the early ages of computer vision. A lot of
research has been conducted and has resulted in many applications. However,
while many segmentation algorithms exist, yet there are only a few sparse and
outdated summarizations available, an overview of the recent achievements and
issues is lacking. We aim to provide a comprehensive review of the recent
progress in this field. Covering 180 publications, we give an overview of broad
areas of segmentation topics including not only the classic bottom-up
approaches, but also the recent development in superpixel, interactive methods,
object proposals, semantic image parsing and image cosegmentation. In addition,
we also review the existing influential datasets and evaluation metrics.
Finally, we suggest some design flavors and research directions for future
research in image segmentation.
| [
{
"version": "v1",
"created": "Tue, 3 Feb 2015 03:00:52 GMT"
}
] | 2015-02-04T00:00:00 | [
[
"Zhu",
"Hongyuan",
""
],
[
"Meng",
"Fanman",
""
],
[
"Cai",
"Jianfei",
""
],
[
"Lu",
"Shijian",
""
]
] | TITLE: Beyond Pixels: A Comprehensive Survey from Bottom-up to Semantic Image
Segmentation and Cosegmentation
ABSTRACT: Image segmentation refers to the process to divide an image into
nonoverlapping meaningful regions according to human perception, which has
become a classic topic since the early ages of computer vision. A lot of
research has been conducted and has resulted in many applications. However,
while many segmentation algorithms exist, yet there are only a few sparse and
outdated summarizations available, an overview of the recent achievements and
issues is lacking. We aim to provide a comprehensive review of the recent
progress in this field. Covering 180 publications, we give an overview of broad
areas of segmentation topics including not only the classic bottom-up
approaches, but also the recent development in superpixel, interactive methods,
object proposals, semantic image parsing and image cosegmentation. In addition,
we also review the existing influential datasets and evaluation metrics.
Finally, we suggest some design flavors and research directions for future
research in image segmentation.
|
1502.00725 | Hongwei Li | Hongwei Li and Qiang Liu | Cheaper and Better: Selecting Good Workers for Crowdsourcing | null | null | null | null | stat.ML cs.AI cs.LG stat.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Crowdsourcing provides a popular paradigm for data collection at scale. We
study the problem of selecting subsets of workers from a given worker pool to
maximize the accuracy under a budget constraint. One natural question is
whether we should hire as many workers as the budget allows, or restrict on a
small number of top-quality workers. By theoretically analyzing the error rate
of a typical setting in crowdsourcing, we frame the worker selection problem
into a combinatorial optimization problem and propose an algorithm to solve it
efficiently. Empirical results on both simulated and real-world datasets show
that our algorithm is able to select a small number of high-quality workers,
and performs as good as, sometimes even better than, the much larger crowds as
the budget allows.
| [
{
"version": "v1",
"created": "Tue, 3 Feb 2015 03:45:48 GMT"
}
] | 2015-02-04T00:00:00 | [
[
"Li",
"Hongwei",
""
],
[
"Liu",
"Qiang",
""
]
] | TITLE: Cheaper and Better: Selecting Good Workers for Crowdsourcing
ABSTRACT: Crowdsourcing provides a popular paradigm for data collection at scale. We
study the problem of selecting subsets of workers from a given worker pool to
maximize the accuracy under a budget constraint. One natural question is
whether we should hire as many workers as the budget allows, or restrict on a
small number of top-quality workers. By theoretically analyzing the error rate
of a typical setting in crowdsourcing, we frame the worker selection problem
into a combinatorial optimization problem and propose an algorithm to solve it
efficiently. Empirical results on both simulated and real-world datasets show
that our algorithm is able to select a small number of high-quality workers,
and performs as good as, sometimes even better than, the much larger crowds as
the budget allows.
|
1502.00739 | Liang Lin | Wei Yang, Ping Luo, Liang Lin | Clothing Co-Parsing by Joint Image Segmentation and Labeling | 8 pages, 5 figures, CVPR 2014 | Computer Vision and Pattern Recognition (CVPR), 2014 IEEE
Conference on , vol., no., pp.3182,3189, 23-28 June 2014 | 10.1109/CVPR.2014.407 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper aims at developing an integrated system of clothing co-parsing, in
order to jointly parse a set of clothing images (unsegmented but annotated with
tags) into semantic configurations. We propose a data-driven framework
consisting of two phases of inference. The first phase, referred as "image
co-segmentation", iterates to extract consistent regions on images and jointly
refines the regions over all images by employing the exemplar-SVM (E-SVM)
technique [23]. In the second phase (i.e. "region co-labeling"), we construct a
multi-image graphical model by taking the segmented regions as vertices, and
incorporate several contexts of clothing configuration (e.g., item location and
mutual interactions). The joint label assignment can be solved using the
efficient Graph Cuts algorithm. In addition to evaluate our framework on the
Fashionista dataset [30], we construct a dataset called CCP consisting of 2098
high-resolution street fashion photos to demonstrate the performance of our
system. We achieve 90.29% / 88.23% segmentation accuracy and 65.52% / 63.89%
recognition rate on the Fashionista and the CCP datasets, respectively, which
are superior compared with state-of-the-art methods.
| [
{
"version": "v1",
"created": "Tue, 3 Feb 2015 04:59:41 GMT"
}
] | 2015-02-04T00:00:00 | [
[
"Yang",
"Wei",
""
],
[
"Luo",
"Ping",
""
],
[
"Lin",
"Liang",
""
]
] | TITLE: Clothing Co-Parsing by Joint Image Segmentation and Labeling
ABSTRACT: This paper aims at developing an integrated system of clothing co-parsing, in
order to jointly parse a set of clothing images (unsegmented but annotated with
tags) into semantic configurations. We propose a data-driven framework
consisting of two phases of inference. The first phase, referred as "image
co-segmentation", iterates to extract consistent regions on images and jointly
refines the regions over all images by employing the exemplar-SVM (E-SVM)
technique [23]. In the second phase (i.e. "region co-labeling"), we construct a
multi-image graphical model by taking the segmented regions as vertices, and
incorporate several contexts of clothing configuration (e.g., item location and
mutual interactions). The joint label assignment can be solved using the
efficient Graph Cuts algorithm. In addition to evaluate our framework on the
Fashionista dataset [30], we construct a dataset called CCP consisting of 2098
high-resolution street fashion photos to demonstrate the performance of our
system. We achieve 90.29% / 88.23% segmentation accuracy and 65.52% / 63.89%
recognition rate on the Fashionista and the CCP datasets, respectively, which
are superior compared with state-of-the-art methods.
|
1502.00750 | Liang Lin | Xiaodan Liang, Qingxing Cao, Rui Huang, Liang Lin | Recognizing Focal Liver Lesions in Contrast-Enhanced Ultrasound with
Discriminatively Trained Spatio-Temporal Model | 5 pages, 1 figures | Biomedical Imaging (ISBI), 2014 IEEE 11th International Symposium
on , vol., no., pp.1184-1187, April 2014 | 10.1109/ISBI.2014.6868087 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The aim of this study is to provide an automatic computational framework to
assist clinicians in diagnosing Focal Liver Lesions (FLLs) in
Contrast-Enhancement Ultrasound (CEUS). We represent FLLs in a CEUS video clip
as an ensemble of Region-of-Interests (ROIs), whose locations are modeled as
latent variables in a discriminative model. Different types of FLLs are
characterized by both spatial and temporal enhancement patterns of the ROIs.
The model is learned by iteratively inferring the optimal ROI locations and
optimizing the model parameters. To efficiently search the optimal spatial and
temporal locations of the ROIs, we propose a data-driven inference algorithm by
combining effective spatial and temporal pruning. The experiments show that our
method achieves promising results on the largest dataset in the literature (to
the best of our knowledge), which we have made publicly available.
| [
{
"version": "v1",
"created": "Tue, 3 Feb 2015 06:14:30 GMT"
}
] | 2015-02-04T00:00:00 | [
[
"Liang",
"Xiaodan",
""
],
[
"Cao",
"Qingxing",
""
],
[
"Huang",
"Rui",
""
],
[
"Lin",
"Liang",
""
]
] | TITLE: Recognizing Focal Liver Lesions in Contrast-Enhanced Ultrasound with
Discriminatively Trained Spatio-Temporal Model
ABSTRACT: The aim of this study is to provide an automatic computational framework to
assist clinicians in diagnosing Focal Liver Lesions (FLLs) in
Contrast-Enhancement Ultrasound (CEUS). We represent FLLs in a CEUS video clip
as an ensemble of Region-of-Interests (ROIs), whose locations are modeled as
latent variables in a discriminative model. Different types of FLLs are
characterized by both spatial and temporal enhancement patterns of the ROIs.
The model is learned by iteratively inferring the optimal ROI locations and
optimizing the model parameters. To efficiently search the optimal spatial and
temporal locations of the ROIs, we propose a data-driven inference algorithm by
combining effective spatial and temporal pruning. The experiments show that our
method achieves promising results on the largest dataset in the literature (to
the best of our knowledge), which we have made publicly available.
|
1403.0388 | Mohammadzaman Zamani | Mohammadzaman Zamani, Hamid Beigy, and Amirreza Shaban | Cascading Randomized Weighted Majority: A New Online Ensemble Learning
Algorithm | 15 pages, 3 figures | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the increasing volume of data in the world, the best approach for
learning from this data is to exploit an online learning algorithm. Online
ensemble methods are online algorithms which take advantage of an ensemble of
classifiers to predict labels of data. Prediction with expert advice is a
well-studied problem in the online ensemble learning literature. The Weighted
Majority algorithm and the randomized weighted majority (RWM) are the most
well-known solutions to this problem, aiming to converge to the best expert.
Since among some expert, the best one does not necessarily have the minimum
error in all regions of data space, defining specific regions and converging to
the best expert in each of these regions will lead to a better result. In this
paper, we aim to resolve this defect of RWM algorithms by proposing a novel
online ensemble algorithm to the problem of prediction with expert advice. We
propose a cascading version of RWM to achieve not only better experimental
results but also a better error bound for sufficiently large datasets.
| [
{
"version": "v1",
"created": "Mon, 3 Mar 2014 11:05:10 GMT"
},
{
"version": "v2",
"created": "Fri, 5 Dec 2014 17:57:03 GMT"
},
{
"version": "v3",
"created": "Sun, 4 Jan 2015 03:01:38 GMT"
},
{
"version": "v4",
"created": "Mon, 2 Feb 2015 17:18:43 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Zamani",
"Mohammadzaman",
""
],
[
"Beigy",
"Hamid",
""
],
[
"Shaban",
"Amirreza",
""
]
] | TITLE: Cascading Randomized Weighted Majority: A New Online Ensemble Learning
Algorithm
ABSTRACT: With the increasing volume of data in the world, the best approach for
learning from this data is to exploit an online learning algorithm. Online
ensemble methods are online algorithms which take advantage of an ensemble of
classifiers to predict labels of data. Prediction with expert advice is a
well-studied problem in the online ensemble learning literature. The Weighted
Majority algorithm and the randomized weighted majority (RWM) are the most
well-known solutions to this problem, aiming to converge to the best expert.
Since among some expert, the best one does not necessarily have the minimum
error in all regions of data space, defining specific regions and converging to
the best expert in each of these regions will lead to a better result. In this
paper, we aim to resolve this defect of RWM algorithms by proposing a novel
online ensemble algorithm to the problem of prediction with expert advice. We
propose a cascading version of RWM to achieve not only better experimental
results but also a better error bound for sufficiently large datasets.
|
1408.4910 | Kai Zhao | Kai Zhao, Mirco Musolesi, Pan Hui, Weixiong Rao and Sasu Tarkoma | Explaining the Power-law Distribution of Human Mobility Through
Transportation Modality Decomposition | null | null | null | null | physics.soc-ph cs.SI physics.data-an | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Human mobility has been empirically observed to exhibit Levy flight
characteristics and behaviour with power-law distributed jump size. The
fundamental mechanisms behind this behaviour has not yet been fully explained.
In this paper, we analyze urban human mobility and we propose to explain the
Levy walk behaviour observed in human mobility patterns by decomposing them
into different classes according to the different transportation modes, such as
Walk/Run, Bicycle, Train/Subway or Car/Taxi/Bus. Our analysis is based on two
real-life GPS datasets containing approximately 10 and 20 million GPS samples
with transportation mode information. We show that human mobility can be
modelled as a mixture of different transportation modes, and that these single
movement patterns can be approximated by a lognormal distribution rather than a
power-law distribution. Then, we demonstrate that the mixture of the decomposed
lognormal flight distributions associated with each modality is a power-law
distribution, providing an explanation to the emergence of Levy Walk patterns
that characterize human mobility patterns.
| [
{
"version": "v1",
"created": "Thu, 21 Aug 2014 08:19:19 GMT"
},
{
"version": "v2",
"created": "Mon, 8 Sep 2014 09:23:48 GMT"
},
{
"version": "v3",
"created": "Mon, 2 Feb 2015 08:30:34 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Zhao",
"Kai",
""
],
[
"Musolesi",
"Mirco",
""
],
[
"Hui",
"Pan",
""
],
[
"Rao",
"Weixiong",
""
],
[
"Tarkoma",
"Sasu",
""
]
] | TITLE: Explaining the Power-law Distribution of Human Mobility Through
Transportation Modality Decomposition
ABSTRACT: Human mobility has been empirically observed to exhibit Levy flight
characteristics and behaviour with power-law distributed jump size. The
fundamental mechanisms behind this behaviour has not yet been fully explained.
In this paper, we analyze urban human mobility and we propose to explain the
Levy walk behaviour observed in human mobility patterns by decomposing them
into different classes according to the different transportation modes, such as
Walk/Run, Bicycle, Train/Subway or Car/Taxi/Bus. Our analysis is based on two
real-life GPS datasets containing approximately 10 and 20 million GPS samples
with transportation mode information. We show that human mobility can be
modelled as a mixture of different transportation modes, and that these single
movement patterns can be approximated by a lognormal distribution rather than a
power-law distribution. Then, we demonstrate that the mixture of the decomposed
lognormal flight distributions associated with each modality is a power-law
distribution, providing an explanation to the emergence of Levy Walk patterns
that characterize human mobility patterns.
|
1501.07093 | Mallenahalli Naresh Kumar Prof. Dr. | V. Sree Hari Rao and M. Naresh Kumar | Novel Approaches for Predicting Risk Factors of Atherosclerosis | 7 pages, 2 figures | Biomedical and Health Informatics, IEEE Journal of , vol.17, no.1,
pp.183,189, Jan. 2013 | 10.1109/TITB.2012.2227271 | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Coronary heart disease (CHD) caused by hardening of artery walls due to
cholesterol known as atherosclerosis is responsible for large number of deaths
world-wide. The disease progression is slow, asymptomatic and may lead to
sudden cardiac arrest, stroke or myocardial infraction. Presently, imaging
techniques are being employed to understand the molecular and metabolic
activity of atherosclerotic plaques to estimate the risk. Though imaging
methods are able to provide some information on plaque metabolism they lack the
required resolution and sensitivity for detection. In this paper we consider
the clinical observations and habits of individuals for predicting the risk
factors of CHD. The identification of risk factors helps in stratifying
patients for further intensive tests such as nuclear imaging or coronary
angiography. We present a novel approach for predicting the risk factors of
atherosclerosis with an in-built imputation algorithm and particle swarm
optimization (PSO). We compare the performance of our methodology with other
machine learning techniques on STULONG dataset which is based on longitudinal
study of middle aged individuals lasting for twenty years. Our methodology
powered by PSO search has identified physical inactivity as one of the risk
factor for the onset of atherosclerosis in addition to other already known
factors. The decision rules extracted by our methodology are able to predict
the risk factors with an accuracy of $99.73%$ which is higher than the
accuracies obtained by application of the state-of-the-art machine learning
techniques presently being employed in the identification of atherosclerosis
risk studies.
| [
{
"version": "v1",
"created": "Wed, 28 Jan 2015 13:26:02 GMT"
},
{
"version": "v2",
"created": "Sat, 31 Jan 2015 03:22:15 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Rao",
"V. Sree Hari",
""
],
[
"Kumar",
"M. Naresh",
""
]
] | TITLE: Novel Approaches for Predicting Risk Factors of Atherosclerosis
ABSTRACT: Coronary heart disease (CHD) caused by hardening of artery walls due to
cholesterol known as atherosclerosis is responsible for large number of deaths
world-wide. The disease progression is slow, asymptomatic and may lead to
sudden cardiac arrest, stroke or myocardial infraction. Presently, imaging
techniques are being employed to understand the molecular and metabolic
activity of atherosclerotic plaques to estimate the risk. Though imaging
methods are able to provide some information on plaque metabolism they lack the
required resolution and sensitivity for detection. In this paper we consider
the clinical observations and habits of individuals for predicting the risk
factors of CHD. The identification of risk factors helps in stratifying
patients for further intensive tests such as nuclear imaging or coronary
angiography. We present a novel approach for predicting the risk factors of
atherosclerosis with an in-built imputation algorithm and particle swarm
optimization (PSO). We compare the performance of our methodology with other
machine learning techniques on STULONG dataset which is based on longitudinal
study of middle aged individuals lasting for twenty years. Our methodology
powered by PSO search has identified physical inactivity as one of the risk
factor for the onset of atherosclerosis in addition to other already known
factors. The decision rules extracted by our methodology are able to predict
the risk factors with an accuracy of $99.73%$ which is higher than the
accuracies obtained by application of the state-of-the-art machine learning
techniques presently being employed in the identification of atherosclerosis
risk studies.
|
1502.00030 | Sravanthi Bondugula | Sravanthi Bondugula, Varun Manjunatha, Larry S. Davis, David Doermann | SHOE: Supervised Hashing with Output Embeddings | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a supervised binary encoding scheme for image retrieval that
learns projections by taking into account similarity between classes obtained
from output embeddings. Our motivation is that binary hash codes learned in
this way improve both the visual quality of retrieval results and existing
supervised hashing schemes. We employ a sequential greedy optimization that
learns relationship aware projections by minimizing the difference between
inner products of binary codes and output embedding vectors. We develop a joint
optimization framework to learn projections which improve the accuracy of
supervised hashing over the current state of the art with respect to standard
and sibling evaluation metrics. We further boost performance by applying the
supervised dimensionality reduction technique on kernelized input CNN features.
Experiments are performed on three datasets: CUB-2011, SUN-Attribute and
ImageNet ILSVRC 2010. As a by-product of our method, we show that using a
simple k-nn pooling classifier with our discriminative codes improves over the
complex classification models on fine grained datasets like CUB and offer an
impressive compression ratio of 1024 on CNN features.
| [
{
"version": "v1",
"created": "Fri, 30 Jan 2015 22:04:12 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Bondugula",
"Sravanthi",
""
],
[
"Manjunatha",
"Varun",
""
],
[
"Davis",
"Larry S.",
""
],
[
"Doermann",
"David",
""
]
] | TITLE: SHOE: Supervised Hashing with Output Embeddings
ABSTRACT: We present a supervised binary encoding scheme for image retrieval that
learns projections by taking into account similarity between classes obtained
from output embeddings. Our motivation is that binary hash codes learned in
this way improve both the visual quality of retrieval results and existing
supervised hashing schemes. We employ a sequential greedy optimization that
learns relationship aware projections by minimizing the difference between
inner products of binary codes and output embedding vectors. We develop a joint
optimization framework to learn projections which improve the accuracy of
supervised hashing over the current state of the art with respect to standard
and sibling evaluation metrics. We further boost performance by applying the
supervised dimensionality reduction technique on kernelized input CNN features.
Experiments are performed on three datasets: CUB-2011, SUN-Attribute and
ImageNet ILSVRC 2010. As a by-product of our method, we show that using a
simple k-nn pooling classifier with our discriminative codes improves over the
complex classification models on fine grained datasets like CUB and offer an
impressive compression ratio of 1024 on CNN features.
|
1502.00046 | Davis King | Davis E. King | Max-Margin Object Detection | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most object detection methods operate by applying a binary classifier to
sub-windows of an image, followed by a non-maximum suppression step where
detections on overlapping sub-windows are removed. Since the number of possible
sub-windows in even moderately sized image datasets is extremely large, the
classifier is typically learned from only a subset of the windows. This avoids
the computational difficulty of dealing with the entire set of sub-windows,
however, as we will show in this paper, it leads to sub-optimal detector
performance.
In particular, the main contribution of this paper is the introduction of a
new method, Max-Margin Object Detection (MMOD), for learning to detect objects
in images. This method does not perform any sub-sampling, but instead optimizes
over all sub-windows. MMOD can be used to improve any object detection method
which is linear in the learned parameters, such as HOG or bag-of-visual-word
models. Using this approach we show substantial performance gains on three
publicly available datasets. Strikingly, we show that a single rigid HOG filter
can outperform a state-of-the-art deformable part model on the Face Detection
Data Set and Benchmark when the HOG filter is learned via MMOD.
| [
{
"version": "v1",
"created": "Sat, 31 Jan 2015 00:32:34 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"King",
"Davis E.",
""
]
] | TITLE: Max-Margin Object Detection
ABSTRACT: Most object detection methods operate by applying a binary classifier to
sub-windows of an image, followed by a non-maximum suppression step where
detections on overlapping sub-windows are removed. Since the number of possible
sub-windows in even moderately sized image datasets is extremely large, the
classifier is typically learned from only a subset of the windows. This avoids
the computational difficulty of dealing with the entire set of sub-windows,
however, as we will show in this paper, it leads to sub-optimal detector
performance.
In particular, the main contribution of this paper is the introduction of a
new method, Max-Margin Object Detection (MMOD), for learning to detect objects
in images. This method does not perform any sub-sampling, but instead optimizes
over all sub-windows. MMOD can be used to improve any object detection method
which is linear in the learned parameters, such as HOG or bag-of-visual-word
models. Using this approach we show substantial performance gains on three
publicly available datasets. Strikingly, we show that a single rigid HOG filter
can outperform a state-of-the-art deformable part model on the Face Detection
Data Set and Benchmark when the HOG filter is learned via MMOD.
|
1502.00093 | Sotetsu Koyamada | Sotetsu Koyamada and Yumi Shikauchi and Ken Nakae and Masanori Koyama
and Shin Ishii | Deep learning of fMRI big data: a novel approach to subject-transfer
decoding | null | null | null | null | stat.ML cs.LG q-bio.NC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As a technology to read brain states from measurable brain activities, brain
decoding are widely applied in industries and medical sciences. In spite of
high demands in these applications for a universal decoder that can be applied
to all individuals simultaneously, large variation in brain activities across
individuals has limited the scope of many studies to the development of
individual-specific decoders. In this study, we used deep neural network (DNN),
a nonlinear hierarchical model, to construct a subject-transfer decoder. Our
decoder is the first successful DNN-based subject-transfer decoder. When
applied to a large-scale functional magnetic resonance imaging (fMRI) database,
our DNN-based decoder achieved higher decoding accuracy than other baseline
methods, including support vector machine (SVM). In order to analyze the
knowledge acquired by this decoder, we applied principal sensitivity analysis
(PSA) to the decoder and visualized the discriminative features that are common
to all subjects in the dataset. Our PSA successfully visualized the
subject-independent features contributing to the subject-transferability of the
trained decoder.
| [
{
"version": "v1",
"created": "Sat, 31 Jan 2015 11:58:26 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Koyamada",
"Sotetsu",
""
],
[
"Shikauchi",
"Yumi",
""
],
[
"Nakae",
"Ken",
""
],
[
"Koyama",
"Masanori",
""
],
[
"Ishii",
"Shin",
""
]
] | TITLE: Deep learning of fMRI big data: a novel approach to subject-transfer
decoding
ABSTRACT: As a technology to read brain states from measurable brain activities, brain
decoding are widely applied in industries and medical sciences. In spite of
high demands in these applications for a universal decoder that can be applied
to all individuals simultaneously, large variation in brain activities across
individuals has limited the scope of many studies to the development of
individual-specific decoders. In this study, we used deep neural network (DNN),
a nonlinear hierarchical model, to construct a subject-transfer decoder. Our
decoder is the first successful DNN-based subject-transfer decoder. When
applied to a large-scale functional magnetic resonance imaging (fMRI) database,
our DNN-based decoder achieved higher decoding accuracy than other baseline
methods, including support vector machine (SVM). In order to analyze the
knowledge acquired by this decoder, we applied principal sensitivity analysis
(PSA) to the decoder and visualized the discriminative features that are common
to all subjects in the dataset. Our PSA successfully visualized the
subject-independent features contributing to the subject-transferability of the
trained decoder.
|
1502.00141 | Mathieu Lagrange | Mathieu Lagrange, Gr\'egoire Lafay, Mathias Rossignol, Emmanouil
Benetos, Axel Roebel | An evaluation framework for event detection using a morphological model
of acoustic scenes | null | null | null | null | stat.ML cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces a model of environmental acoustic scenes which adopts a
morphological approach by ab-stracting temporal structures of acoustic scenes.
To demonstrate its potential, this model is employed to evaluate the
performance of a large set of acoustic events detection systems. This model
allows us to explicitly control key morphological aspects of the acoustic scene
and isolate their impact on the performance of the system under evaluation.
Thus, more information can be gained on the behavior of evaluated systems,
providing guidance for further improvements. The proposed model is validated
using submitted systems from the IEEE DCASE Challenge; results indicate that
the proposed scheme is able to successfully build datasets useful for
evaluating some aspects the performance of event detection systems, more
particularly their robustness to new listening conditions and the increasing
level of background sounds.
| [
{
"version": "v1",
"created": "Sat, 31 Jan 2015 18:12:34 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Lagrange",
"Mathieu",
""
],
[
"Lafay",
"Grégoire",
""
],
[
"Rossignol",
"Mathias",
""
],
[
"Benetos",
"Emmanouil",
""
],
[
"Roebel",
"Axel",
""
]
] | TITLE: An evaluation framework for event detection using a morphological model
of acoustic scenes
ABSTRACT: This paper introduces a model of environmental acoustic scenes which adopts a
morphological approach by ab-stracting temporal structures of acoustic scenes.
To demonstrate its potential, this model is employed to evaluate the
performance of a large set of acoustic events detection systems. This model
allows us to explicitly control key morphological aspects of the acoustic scene
and isolate their impact on the performance of the system under evaluation.
Thus, more information can be gained on the behavior of evaluated systems,
providing guidance for further improvements. The proposed model is validated
using submitted systems from the IEEE DCASE Challenge; results indicate that
the proposed scheme is able to successfully build datasets useful for
evaluating some aspects the performance of event detection systems, more
particularly their robustness to new listening conditions and the increasing
level of background sounds.
|
1502.00166 | Marijn ten Thij | Marijn ten Thij, Tanneke Ouboter, Daniel Worm, Nelly Litvak, Hans van
den Berg and Sandjai Bhulai | Modelling of trends in Twitter using retweet graph dynamics | 16 pages, 5 figures, presented at WAW 2014 | Algorithms and Models for the Web Graph, 11th International
Workshop, WAW 2014, Beijing, China, December 17-18, 2014, Proceedings pp
132-147, Lecture Notes in Computer Science, Springer | 10.1007/978-3-319-13123-8_11 | null | cs.SI physics.soc-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we model user behaviour in Twitter to capture the emergence of
trending topics. For this purpose, we first extensively analyse tweet datasets
of several different events. In particular, for these datasets, we construct
and investigate the retweet graphs. We find that the retweet graph for a
trending topic has a relatively dense largest connected component (LCC). Next,
based on the insights obtained from the analyses of the datasets, we design a
mathematical model that describes the evolution of a retweet graph by three
main parameters. We then quantify, analytically and by simulation, the
influence of the model parameters on the basic characteristics of the retweet
graph, such as the density of edges and the size and density of the LCC.
Finally, we put the model in practice, estimate its parameters and compare the
resulting behavior of the model to our datasets.
| [
{
"version": "v1",
"created": "Sat, 31 Jan 2015 21:45:17 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Thij",
"Marijn ten",
""
],
[
"Ouboter",
"Tanneke",
""
],
[
"Worm",
"Daniel",
""
],
[
"Litvak",
"Nelly",
""
],
[
"Berg",
"Hans van den",
""
],
[
"Bhulai",
"Sandjai",
""
]
] | TITLE: Modelling of trends in Twitter using retweet graph dynamics
ABSTRACT: In this paper we model user behaviour in Twitter to capture the emergence of
trending topics. For this purpose, we first extensively analyse tweet datasets
of several different events. In particular, for these datasets, we construct
and investigate the retweet graphs. We find that the retweet graph for a
trending topic has a relatively dense largest connected component (LCC). Next,
based on the insights obtained from the analyses of the datasets, we design a
mathematical model that describes the evolution of a retweet graph by three
main parameters. We then quantify, analytically and by simulation, the
influence of the model parameters on the basic characteristics of the retweet
graph, such as the density of edges and the size and density of the LCC.
Finally, we put the model in practice, estimate its parameters and compare the
resulting behavior of the model to our datasets.
|
1502.00192 | Menglong Zhu | Menglong Zhu, Xiaowei Zhou, Kostas Daniilidis | Pose and Shape Estimation with Discriminatively Learned Parts | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a new approach for estimating the 3D pose and the 3D shape of an
object from a single image. Given a training set of view exemplars, we learn
and select appearance-based discriminative parts which are mapped onto the 3D
model from the training set through a facil- ity location optimization. The
training set of 3D models is summarized into a sparse set of shapes from which
we can generalize by linear combination. Given a test picture, we detect
hypotheses for each part. The main challenge is to select from these hypotheses
and compute the 3D pose and shape coefficients at the same time. To achieve
this, we optimize a function that minimizes simultaneously the geometric
reprojection error as well as the appearance matching of the parts. We apply
the alternating direction method of multipliers (ADMM) to minimize the
resulting convex function. We evaluate our approach on the Fine Grained 3D Car
dataset with superior performance in shape and pose errors. Our main and novel
contribution is the simultaneous solution for part localization, 3D pose and
shape by maximizing both geometric and appearance compatibility.
| [
{
"version": "v1",
"created": "Sun, 1 Feb 2015 04:09:23 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Zhu",
"Menglong",
""
],
[
"Zhou",
"Xiaowei",
""
],
[
"Daniilidis",
"Kostas",
""
]
] | TITLE: Pose and Shape Estimation with Discriminatively Learned Parts
ABSTRACT: We introduce a new approach for estimating the 3D pose and the 3D shape of an
object from a single image. Given a training set of view exemplars, we learn
and select appearance-based discriminative parts which are mapped onto the 3D
model from the training set through a facil- ity location optimization. The
training set of 3D models is summarized into a sparse set of shapes from which
we can generalize by linear combination. Given a test picture, we detect
hypotheses for each part. The main challenge is to select from these hypotheses
and compute the 3D pose and shape coefficients at the same time. To achieve
this, we optimize a function that minimizes simultaneously the geometric
reprojection error as well as the appearance matching of the parts. We apply
the alternating direction method of multipliers (ADMM) to minimize the
resulting convex function. We evaluate our approach on the Fine Grained 3D Car
dataset with superior performance in shape and pose errors. Our main and novel
contribution is the simultaneous solution for part localization, 3D pose and
shape by maximizing both geometric and appearance compatibility.
|
1502.00194 | James J.Q. Yu | James J.Q. Yu, Albert Y.S. Lam, Victor O.K. Li | Real-Coded Chemical Reaction Optimization with Different Perturbation
Functions | null | null | 10.1109/CEC.2012.6252925 | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Chemical Reaction Optimization (CRO) is a powerful metaheuristic which mimics
the interactions of molecules in chemical reactions to search for the global
optimum. The perturbation function greatly influences the performance of CRO on
solving different continuous problems. In this paper, we study four different
probability distributions, namely, the Gaussian distribution, the Cauchy
distribution, the exponential distribution, and a modified Rayleigh
distribution, for the perturbation function of CRO. Different distributions
have different impacts on the solutions. The distributions are tested by a set
of well-known benchmark functions and simulation results show that problems
with different characteristics have different preference on the distribution
function. Our study gives guidelines to design CRO for different types of
optimization problems.
| [
{
"version": "v1",
"created": "Sun, 1 Feb 2015 04:45:43 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Yu",
"James J. Q.",
""
],
[
"Lam",
"Albert Y. S.",
""
],
[
"Li",
"Victor O. K.",
""
]
] | TITLE: Real-Coded Chemical Reaction Optimization with Different Perturbation
Functions
ABSTRACT: Chemical Reaction Optimization (CRO) is a powerful metaheuristic which mimics
the interactions of molecules in chemical reactions to search for the global
optimum. The perturbation function greatly influences the performance of CRO on
solving different continuous problems. In this paper, we study four different
probability distributions, namely, the Gaussian distribution, the Cauchy
distribution, the exponential distribution, and a modified Rayleigh
distribution, for the perturbation function of CRO. Different distributions
have different impacts on the solutions. The distributions are tested by a set
of well-known benchmark functions and simulation results show that problems
with different characteristics have different preference on the distribution
function. Our study gives guidelines to design CRO for different types of
optimization problems.
|
1502.00197 | James J.Q. Yu | James J.Q. Yu, Victor O.K. Li, Albert Y.S. Lam | An Inter-molecular Adaptive Collision Scheme for Chemical Reaction
Optimization | null | null | 10.1109/CEC.2014.6900234 | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Optimization techniques are frequently applied in science and engineering
research and development. Evolutionary algorithms, as a kind of general-purpose
metaheuristic, have been shown to be very effective in solving a wide range of
optimization problems. A recently proposed chemical-reaction-inspired
metaheuristic, Chemical Reaction Optimization (CRO), has been applied to solve
many global optimization problems. However, the functionality of the
inter-molecular ineffective collision operator in the canonical CRO design
overlaps that of the on-wall ineffective collision operator, which can
potential impair the overall performance. In this paper we propose a new
inter-molecular ineffective collision operator for CRO for global optimization.
To fully utilize our newly proposed operator, we also design a scheme to adapt
the algorithm to optimization problems with different search space
characteristics. We analyze the performance of our proposed algorithm with a
number of widely used benchmark functions. The simulation results indicate that
the new algorithm has superior performance over the canonical CRO.
| [
{
"version": "v1",
"created": "Sun, 1 Feb 2015 04:53:48 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Yu",
"James J. Q.",
""
],
[
"Li",
"Victor O. K.",
""
],
[
"Lam",
"Albert Y. S.",
""
]
] | TITLE: An Inter-molecular Adaptive Collision Scheme for Chemical Reaction
Optimization
ABSTRACT: Optimization techniques are frequently applied in science and engineering
research and development. Evolutionary algorithms, as a kind of general-purpose
metaheuristic, have been shown to be very effective in solving a wide range of
optimization problems. A recently proposed chemical-reaction-inspired
metaheuristic, Chemical Reaction Optimization (CRO), has been applied to solve
many global optimization problems. However, the functionality of the
inter-molecular ineffective collision operator in the canonical CRO design
overlaps that of the on-wall ineffective collision operator, which can
potential impair the overall performance. In this paper we propose a new
inter-molecular ineffective collision operator for CRO for global optimization.
To fully utilize our newly proposed operator, we also design a scheme to adapt
the algorithm to optimization problems with different search space
characteristics. We analyze the performance of our proposed algorithm with a
number of widely used benchmark functions. The simulation results indicate that
the new algorithm has superior performance over the canonical CRO.
|
1502.00199 | James J.Q. Yu | James J.Q. Yu, Albert Y.S. Lam, Victor O.K. Li | Chemical Reaction Optimization for the Set Covering Problem | null | null | 10.1109/CEC.2014.6900233 | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The set covering problem (SCP) is one of the representative combinatorial
optimization problems, having many practical applications. This paper
investigates the development of an algorithm to solve SCP by employing chemical
reaction optimization (CRO), a general-purpose metaheuristic. It is tested on a
wide range of benchmark instances of SCP. The simulation results indicate that
this algorithm gives outstanding performance compared with other heuristics and
metaheuristics in solving SCP.
| [
{
"version": "v1",
"created": "Sun, 1 Feb 2015 04:56:13 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Yu",
"James J. Q.",
""
],
[
"Lam",
"Albert Y. S.",
""
],
[
"Li",
"Victor O. K.",
""
]
] | TITLE: Chemical Reaction Optimization for the Set Covering Problem
ABSTRACT: The set covering problem (SCP) is one of the representative combinatorial
optimization problems, having many practical applications. This paper
investigates the development of an algorithm to solve SCP by employing chemical
reaction optimization (CRO), a general-purpose metaheuristic. It is tested on a
wide range of benchmark instances of SCP. The simulation results indicate that
this algorithm gives outstanding performance compared with other heuristics and
metaheuristics in solving SCP.
|
1502.00231 | Yishi Zhang | Zhijun Chen, Chaozhong Wu, Yishi Zhang, Zhen Huang, Bin Ran, Ming
Zhong, Nengchao Lyu | Feature Selection with Redundancy-complementariness Dispersion | 28 pages, 13 figures, 7 tables | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Feature selection has attracted significant attention in data mining and
machine learning in the past decades. Many existing feature selection methods
eliminate redundancy by measuring pairwise inter-correlation of features,
whereas the complementariness of features and higher inter-correlation among
more than two features are ignored. In this study, a modification item
concerning the complementariness of features is introduced in the evaluation
criterion of features. Additionally, in order to identify the interference
effect of already-selected False Positives (FPs), the
redundancy-complementariness dispersion is also taken into account to adjust
the measurement of pairwise inter-correlation of features. To illustrate the
effectiveness of proposed method, classification experiments are applied with
four frequently used classifiers on ten datasets. Classification results verify
the superiority of proposed method compared with five representative feature
selection methods.
| [
{
"version": "v1",
"created": "Sun, 1 Feb 2015 10:44:26 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Chen",
"Zhijun",
""
],
[
"Wu",
"Chaozhong",
""
],
[
"Zhang",
"Yishi",
""
],
[
"Huang",
"Zhen",
""
],
[
"Ran",
"Bin",
""
],
[
"Zhong",
"Ming",
""
],
[
"Lyu",
"Nengchao",
""
]
] | TITLE: Feature Selection with Redundancy-complementariness Dispersion
ABSTRACT: Feature selection has attracted significant attention in data mining and
machine learning in the past decades. Many existing feature selection methods
eliminate redundancy by measuring pairwise inter-correlation of features,
whereas the complementariness of features and higher inter-correlation among
more than two features are ignored. In this study, a modification item
concerning the complementariness of features is introduced in the evaluation
criterion of features. Additionally, in order to identify the interference
effect of already-selected False Positives (FPs), the
redundancy-complementariness dispersion is also taken into account to adjust
the measurement of pairwise inter-correlation of features. To illustrate the
effectiveness of proposed method, classification experiments are applied with
four frequently used classifiers on ten datasets. Classification results verify
the superiority of proposed method compared with five representative feature
selection methods.
|
1502.00303 | Xianbiao Qi | Xianbiao Qi, Chun-Guang Li, Guoying Zhao, Xiaopeng Hong, Matti
Pietik\"ainen | Dynamic texture and scene classification by transferring deep image
features | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Dynamic texture and scene classification are two fundamental problems in
understanding natural video content. Extracting robust and effective features
is a crucial step towards solving these problems. However the existing
approaches suffer from the sensitivity to either varying illumination, or
viewpoint changing, or even camera motion, and/or the lack of spatial
information. Inspired by the success of deep structures in image
classification, we attempt to leverage a deep structure to extract feature for
dynamic texture and scene classification. To tackle with the challenges in
training a deep structure, we propose to transfer some prior knowledge from
image domain to video domain. To be specific, we propose to apply a
well-trained Convolutional Neural Network (ConvNet) as a mid-level feature
extractor to extract features from each frame, and then form a representation
of a video by concatenating the first and the second order statistics over the
mid-level features. We term this two-level feature extraction scheme as a
Transferred ConvNet Feature (TCoF). Moreover we explore two different
implementations of the TCoF scheme, i.e., the \textit{spatial} TCoF and the
\textit{temporal} TCoF, in which the mean-removed frames and the difference
between two adjacent frames are used as the inputs of the ConvNet,
respectively. We evaluate systematically the proposed spatial TCoF and the
temporal TCoF schemes on three benchmark data sets, including DynTex, YUPENN,
and Maryland, and demonstrate that the proposed approach yields superior
performance.
| [
{
"version": "v1",
"created": "Sun, 1 Feb 2015 20:22:00 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Qi",
"Xianbiao",
""
],
[
"Li",
"Chun-Guang",
""
],
[
"Zhao",
"Guoying",
""
],
[
"Hong",
"Xiaopeng",
""
],
[
"Pietikäinen",
"Matti",
""
]
] | TITLE: Dynamic texture and scene classification by transferring deep image
features
ABSTRACT: Dynamic texture and scene classification are two fundamental problems in
understanding natural video content. Extracting robust and effective features
is a crucial step towards solving these problems. However the existing
approaches suffer from the sensitivity to either varying illumination, or
viewpoint changing, or even camera motion, and/or the lack of spatial
information. Inspired by the success of deep structures in image
classification, we attempt to leverage a deep structure to extract feature for
dynamic texture and scene classification. To tackle with the challenges in
training a deep structure, we propose to transfer some prior knowledge from
image domain to video domain. To be specific, we propose to apply a
well-trained Convolutional Neural Network (ConvNet) as a mid-level feature
extractor to extract features from each frame, and then form a representation
of a video by concatenating the first and the second order statistics over the
mid-level features. We term this two-level feature extraction scheme as a
Transferred ConvNet Feature (TCoF). Moreover we explore two different
implementations of the TCoF scheme, i.e., the \textit{spatial} TCoF and the
\textit{temporal} TCoF, in which the mean-removed frames and the difference
between two adjacent frames are used as the inputs of the ConvNet,
respectively. We evaluate systematically the proposed spatial TCoF and the
temporal TCoF schemes on three benchmark data sets, including DynTex, YUPENN,
and Maryland, and demonstrate that the proposed approach yields superior
performance.
|
1502.00341 | Liang Lin | Liang Lin, Xiaolong Wang, Wei Yang, Jian-Huang Lai | Discriminatively Trained And-Or Graph Models for Object Shape Detection | 15 pages, 14 figures, TPAMI 2014 | Pattern Analysis and Machine Intelligence, IEEE Transactions on ,
vol.PP, no.99, pp.1,1, 2014 | 10.1109/TPAMI.2014.2359888 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we investigate a novel reconfigurable part-based model, namely
And-Or graph model, to recognize object shapes in images. Our proposed model
consists of four layers: leaf-nodes at the bottom are local classifiers for
detecting contour fragments; or-nodes above the leaf-nodes function as the
switches to activate their child leaf-nodes, making the model reconfigurable
during inference; and-nodes in a higher layer capture holistic shape
deformations; one root-node on the top, which is also an or-node, activates one
of its child and-nodes to deal with large global variations (e.g. different
poses and views). We propose a novel structural optimization algorithm to
discriminatively train the And-Or model from weakly annotated data. This
algorithm iteratively determines the model structures (e.g. the nodes and their
layouts) along with the parameter learning. On several challenging datasets,
our model demonstrates the effectiveness to perform robust shape-based object
detection against background clutter and outperforms the other state-of-the-art
approaches. We also release a new shape database with annotations, which
includes more than 1500 challenging shape instances, for recognition and
detection.
| [
{
"version": "v1",
"created": "Mon, 2 Feb 2015 02:04:01 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Lin",
"Liang",
""
],
[
"Wang",
"Xiaolong",
""
],
[
"Yang",
"Wei",
""
],
[
"Lai",
"Jian-Huang",
""
]
] | TITLE: Discriminatively Trained And-Or Graph Models for Object Shape Detection
ABSTRACT: In this paper, we investigate a novel reconfigurable part-based model, namely
And-Or graph model, to recognize object shapes in images. Our proposed model
consists of four layers: leaf-nodes at the bottom are local classifiers for
detecting contour fragments; or-nodes above the leaf-nodes function as the
switches to activate their child leaf-nodes, making the model reconfigurable
during inference; and-nodes in a higher layer capture holistic shape
deformations; one root-node on the top, which is also an or-node, activates one
of its child and-nodes to deal with large global variations (e.g. different
poses and views). We propose a novel structural optimization algorithm to
discriminatively train the And-Or model from weakly annotated data. This
algorithm iteratively determines the model structures (e.g. the nodes and their
layouts) along with the parameter learning. On several challenging datasets,
our model demonstrates the effectiveness to perform robust shape-based object
detection against background clutter and outperforms the other state-of-the-art
approaches. We also release a new shape database with annotations, which
includes more than 1500 challenging shape instances, for recognition and
detection.
|
1502.00363 | Wangmeng Zuo | Wangmeng Zuo, Faqiang Wang, David Zhang, Liang Lin, Yuchi Huang, Deyu
Meng, Lei Zhang | Iterated Support Vector Machines for Distance Metric Learning | 14 pages, 10 figures | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Distance metric learning aims to learn from the given training data a valid
distance metric, with which the similarity between data samples can be more
effectively evaluated for classification. Metric learning is often formulated
as a convex or nonconvex optimization problem, while many existing metric
learning algorithms become inefficient for large scale problems. In this paper,
we formulate metric learning as a kernel classification problem, and solve it
by iterated training of support vector machines (SVM). The new formulation is
easy to implement, efficient in training, and tractable for large-scale
problems. Two novel metric learning models, namely Positive-semidefinite
Constrained Metric Learning (PCML) and Nonnegative-coefficient Constrained
Metric Learning (NCML), are developed. Both PCML and NCML can guarantee the
global optimality of their solutions. Experimental results on UCI dataset
classification, handwritten digit recognition, face verification and person
re-identification demonstrate that the proposed metric learning methods achieve
higher classification accuracy than state-of-the-art methods and they are
significantly more efficient in training.
| [
{
"version": "v1",
"created": "Mon, 2 Feb 2015 05:30:44 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Zuo",
"Wangmeng",
""
],
[
"Wang",
"Faqiang",
""
],
[
"Zhang",
"David",
""
],
[
"Lin",
"Liang",
""
],
[
"Huang",
"Yuchi",
""
],
[
"Meng",
"Deyu",
""
],
[
"Zhang",
"Lei",
""
]
] | TITLE: Iterated Support Vector Machines for Distance Metric Learning
ABSTRACT: Distance metric learning aims to learn from the given training data a valid
distance metric, with which the similarity between data samples can be more
effectively evaluated for classification. Metric learning is often formulated
as a convex or nonconvex optimization problem, while many existing metric
learning algorithms become inefficient for large scale problems. In this paper,
we formulate metric learning as a kernel classification problem, and solve it
by iterated training of support vector machines (SVM). The new formulation is
easy to implement, efficient in training, and tractable for large-scale
problems. Two novel metric learning models, namely Positive-semidefinite
Constrained Metric Learning (PCML) and Nonnegative-coefficient Constrained
Metric Learning (NCML), are developed. Both PCML and NCML can guarantee the
global optimality of their solutions. Experimental results on UCI dataset
classification, handwritten digit recognition, face verification and person
re-identification demonstrate that the proposed metric learning methods achieve
higher classification accuracy than state-of-the-art methods and they are
significantly more efficient in training.
|
1502.00377 | Liang Lin | Liang Lin, Yongyi Lu, Yan Pan, Xiaowu Chen | Integrating Graph Partitioning and Matching for Trajectory Analysis in
Video Surveillance | 10 pages, 12 figures | Image Processing, IEEE Transactions on , vol.21, no.12,
pp.4844-4857, 2012 | 10.1109/TIP.2012.2211373 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In order to track the moving objects in long range against occlusion,
interruption, and background clutter, this paper proposes a unified approach
for global trajectory analysis. Instead of the traditional frame-by-frame
tracking, our method recovers target trajectories based on a short sequence of
video frames, e.g. $15$ frames. We initially calculate a foreground map at each
frame, as obtained from a state-of-the-art background model. An attribute graph
is then extracted from the foreground map, where the graph vertices are image
primitives represented by the composite features. With this graph
representation, we pose trajectory analysis as a joint task of spatial graph
partitioning and temporal graph matching. The task can be formulated by
maximizing a posteriori under the Bayesian framework, in which we integrate the
spatio-temporal contexts and the appearance models. The probabilistic inference
is achieved by a data-driven Markov Chain Monte Carlo (MCMC) algorithm. Given a
peroid of observed frames, the algorithm simulates a ergodic and aperiodic
Markov Chain, and it visits a sequence of solution states in the joint space of
spatial graph partitioning and temporal graph matching. In the experiments, our
method is tested on several challenging videos from the public datasets of
visual surveillance, and it outperforms the state-of-the-art methods.
| [
{
"version": "v1",
"created": "Mon, 2 Feb 2015 06:52:47 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Lin",
"Liang",
""
],
[
"Lu",
"Yongyi",
""
],
[
"Pan",
"Yan",
""
],
[
"Chen",
"Xiaowu",
""
]
] | TITLE: Integrating Graph Partitioning and Matching for Trajectory Analysis in
Video Surveillance
ABSTRACT: In order to track the moving objects in long range against occlusion,
interruption, and background clutter, this paper proposes a unified approach
for global trajectory analysis. Instead of the traditional frame-by-frame
tracking, our method recovers target trajectories based on a short sequence of
video frames, e.g. $15$ frames. We initially calculate a foreground map at each
frame, as obtained from a state-of-the-art background model. An attribute graph
is then extracted from the foreground map, where the graph vertices are image
primitives represented by the composite features. With this graph
representation, we pose trajectory analysis as a joint task of spatial graph
partitioning and temporal graph matching. The task can be formulated by
maximizing a posteriori under the Bayesian framework, in which we integrate the
spatio-temporal contexts and the appearance models. The probabilistic inference
is achieved by a data-driven Markov Chain Monte Carlo (MCMC) algorithm. Given a
peroid of observed frames, the algorithm simulates a ergodic and aperiodic
Markov Chain, and it visits a sequence of solution states in the joint space of
spatial graph partitioning and temporal graph matching. In the experiments, our
method is tested on several challenging videos from the public datasets of
visual surveillance, and it outperforms the state-of-the-art methods.
|
1502.00416 | Liang Lin | Bo Jiang, Yongyi Lu, Xiying Li, Liang Lin | Towards a solid solution of real-time fire and flame detection | 14 pages, 6 figures | Multimedia Tools and Applications, pp. 1380-7501, 2014 | 10.1007/s11042-014-2106-z | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although the object detection and recognition has received growing attention
for decades, a robust fire and flame detection method is rarely explored. This
paper presents an empirical study, towards a general and solid approach to fast
detect fire and flame in videos, with the applications in video surveillance
and event retrieval. Our system consists of three cascaded steps: (1) candidate
regions proposing by a background model, (2) fire region classifying with
color-texture features and a dictionary of visual words, and (3) temporal
verifying. The experimental evaluation and analysis are done for each step. We
believe that it is a useful service to both academic research and real-world
application. In addition, we release the software of the proposed system with
the source code, as well as a public benchmark and data set, including 64 video
clips covered both indoor and outdoor scenes under different conditions. We
achieve an 82% Recall with 93% Precision on the data set, and greatly improve
the performance by state-of-the-arts methods.
| [
{
"version": "v1",
"created": "Mon, 2 Feb 2015 09:34:01 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Jiang",
"Bo",
""
],
[
"Lu",
"Yongyi",
""
],
[
"Li",
"Xiying",
""
],
[
"Lin",
"Liang",
""
]
] | TITLE: Towards a solid solution of real-time fire and flame detection
ABSTRACT: Although the object detection and recognition has received growing attention
for decades, a robust fire and flame detection method is rarely explored. This
paper presents an empirical study, towards a general and solid approach to fast
detect fire and flame in videos, with the applications in video surveillance
and event retrieval. Our system consists of three cascaded steps: (1) candidate
regions proposing by a background model, (2) fire region classifying with
color-texture features and a dictionary of visual words, and (3) temporal
verifying. The experimental evaluation and analysis are done for each step. We
believe that it is a useful service to both academic research and real-world
application. In addition, we release the software of the proposed system with
the source code, as well as a public benchmark and data set, including 64 video
clips covered both indoor and outdoor scenes under different conditions. We
achieve an 82% Recall with 93% Precision on the data set, and greatly improve
the performance by state-of-the-arts methods.
|
1502.00512 | Will Williams | Will Williams, Niranjani Prasad, David Mrva, Tom Ash, Tony Robinson | Scaling Recurrent Neural Network Language Models | null | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper investigates the scaling properties of Recurrent Neural Network
Language Models (RNNLMs). We discuss how to train very large RNNs on GPUs and
address the questions of how RNNLMs scale with respect to model size,
training-set size, computational costs and memory. Our analysis shows that
despite being more costly to train, RNNLMs obtain much lower perplexities on
standard benchmarks than n-gram models. We train the largest known RNNs and
present relative word error rates gains of 18% on an ASR task. We also present
the new lowest perplexities on the recently released billion word language
modelling benchmark, 1 BLEU point gain on machine translation and a 17%
relative hit rate gain in word prediction.
| [
{
"version": "v1",
"created": "Mon, 2 Feb 2015 15:27:37 GMT"
}
] | 2015-02-03T00:00:00 | [
[
"Williams",
"Will",
""
],
[
"Prasad",
"Niranjani",
""
],
[
"Mrva",
"David",
""
],
[
"Ash",
"Tom",
""
],
[
"Robinson",
"Tony",
""
]
] | TITLE: Scaling Recurrent Neural Network Language Models
ABSTRACT: This paper investigates the scaling properties of Recurrent Neural Network
Language Models (RNNLMs). We discuss how to train very large RNNs on GPUs and
address the questions of how RNNLMs scale with respect to model size,
training-set size, computational costs and memory. Our analysis shows that
despite being more costly to train, RNNLMs obtain much lower perplexities on
standard benchmarks than n-gram models. We train the largest known RNNs and
present relative word error rates gains of 18% on an ASR task. We also present
the new lowest perplexities on the recently released billion word language
modelling benchmark, 1 BLEU point gain on machine translation and a 17%
relative hit rate gain in word prediction.
|
1407.0316 | Mahito Sugiyama | Mahito Sugiyama, Felipe Llinares L\'opez, Niklas Kasenburg, Karsten M.
Borgwardt | Significant Subgraph Mining with Multiple Testing Correction | 18 pages, 5 figure, accepted to the 2015 SIAM International
Conference on Data Mining (SDM15) | null | null | null | stat.ME cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The problem of finding itemsets that are statistically significantly enriched
in a class of transactions is complicated by the need to correct for multiple
hypothesis testing. Pruning untestable hypotheses was recently proposed as a
strategy for this task of significant itemset mining. It was shown to lead to
greater statistical power, the discovery of more truly significant itemsets,
than the standard Bonferroni correction on real-world datasets. An open
question, however, is whether this strategy of excluding untestable hypotheses
also leads to greater statistical power in subgraph mining, in which the number
of hypotheses is much larger than in itemset mining. Here we answer this
question by an empirical investigation on eight popular graph benchmark
datasets. We propose a new efficient search strategy, which always returns the
same solution as the state-of-the-art approach and is approximately two orders
of magnitude faster. Moreover, we exploit the dependence between subgraphs by
considering the effective number of tests and thereby further increase the
statistical power.
| [
{
"version": "v1",
"created": "Tue, 1 Jul 2014 16:53:51 GMT"
},
{
"version": "v2",
"created": "Sun, 6 Jul 2014 13:39:21 GMT"
},
{
"version": "v3",
"created": "Fri, 30 Jan 2015 16:11:17 GMT"
}
] | 2015-02-02T00:00:00 | [
[
"Sugiyama",
"Mahito",
""
],
[
"López",
"Felipe Llinares",
""
],
[
"Kasenburg",
"Niklas",
""
],
[
"Borgwardt",
"Karsten M.",
""
]
] | TITLE: Significant Subgraph Mining with Multiple Testing Correction
ABSTRACT: The problem of finding itemsets that are statistically significantly enriched
in a class of transactions is complicated by the need to correct for multiple
hypothesis testing. Pruning untestable hypotheses was recently proposed as a
strategy for this task of significant itemset mining. It was shown to lead to
greater statistical power, the discovery of more truly significant itemsets,
than the standard Bonferroni correction on real-world datasets. An open
question, however, is whether this strategy of excluding untestable hypotheses
also leads to greater statistical power in subgraph mining, in which the number
of hypotheses is much larger than in itemset mining. Here we answer this
question by an empirical investigation on eight popular graph benchmark
datasets. We propose a new efficient search strategy, which always returns the
same solution as the state-of-the-art approach and is approximately two orders
of magnitude faster. Moreover, we exploit the dependence between subgraphs by
considering the effective number of tests and thereby further increase the
statistical power.
|
1409.0575 | Olga Russakovsky | Olga Russakovsky and Jia Deng and Hao Su and Jonathan Krause and
Sanjeev Satheesh and Sean Ma and Zhiheng Huang and Andrej Karpathy and Aditya
Khosla and Michael Bernstein and Alexander C. Berg and Li Fei-Fei | ImageNet Large Scale Visual Recognition Challenge | 43 pages, 16 figures. v3 includes additional comparisons with PASCAL
VOC (per-category comparisons in Table 3, distribution of localization
difficulty in Fig 16), a list of queries used for obtaining object detection
images (Appendix C), and some additional references | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The ImageNet Large Scale Visual Recognition Challenge is a benchmark in
object category classification and detection on hundreds of object categories
and millions of images. The challenge has been run annually from 2010 to
present, attracting participation from more than fifty institutions.
This paper describes the creation of this benchmark dataset and the advances
in object recognition that have been possible as a result. We discuss the
challenges of collecting large-scale ground truth annotation, highlight key
breakthroughs in categorical object recognition, provide a detailed analysis of
the current state of the field of large-scale image classification and object
detection, and compare the state-of-the-art computer vision accuracy with human
accuracy. We conclude with lessons learned in the five years of the challenge,
and propose future directions and improvements.
| [
{
"version": "v1",
"created": "Mon, 1 Sep 2014 22:29:38 GMT"
},
{
"version": "v2",
"created": "Mon, 1 Dec 2014 01:08:31 GMT"
},
{
"version": "v3",
"created": "Fri, 30 Jan 2015 01:23:59 GMT"
}
] | 2015-02-02T00:00:00 | [
[
"Russakovsky",
"Olga",
""
],
[
"Deng",
"Jia",
""
],
[
"Su",
"Hao",
""
],
[
"Krause",
"Jonathan",
""
],
[
"Satheesh",
"Sanjeev",
""
],
[
"Ma",
"Sean",
""
],
[
"Huang",
"Zhiheng",
""
],
[
"Karpathy",
"Andrej",
""
],
[
"Khosla",
"Aditya",
""
],
[
"Bernstein",
"Michael",
""
],
[
"Berg",
"Alexander C.",
""
],
[
"Fei-Fei",
"Li",
""
]
] | TITLE: ImageNet Large Scale Visual Recognition Challenge
ABSTRACT: The ImageNet Large Scale Visual Recognition Challenge is a benchmark in
object category classification and detection on hundreds of object categories
and millions of images. The challenge has been run annually from 2010 to
present, attracting participation from more than fifty institutions.
This paper describes the creation of this benchmark dataset and the advances
in object recognition that have been possible as a result. We discuss the
challenges of collecting large-scale ground truth annotation, highlight key
breakthroughs in categorical object recognition, provide a detailed analysis of
the current state of the field of large-scale image classification and object
detection, and compare the state-of-the-art computer vision accuracy with human
accuracy. We conclude with lessons learned in the five years of the challenge,
and propose future directions and improvements.
|
1411.6243 | Xu Sun | Xu Sun | Structure Regularization for Structured Prediction: Theories and
Experiments | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While there are many studies on weight regularization, the study on structure
regularization is rare. Many existing systems on structured prediction focus on
increasing the level of structural dependencies within the model. However, this
trend could have been misdirected, because our study suggests that complex
structures are actually harmful to generalization ability in structured
prediction. To control structure-based overfitting, we propose a structure
regularization framework via \emph{structure decomposition}, which decomposes
training samples into mini-samples with simpler structures, deriving a model
with better generalization power. We show both theoretically and empirically
that structure regularization can effectively control overfitting risk and lead
to better accuracy. As a by-product, the proposed method can also substantially
accelerate the training speed. The method and the theoretical results can apply
to general graphical models with arbitrary structures. Experiments on
well-known tasks demonstrate that our method can easily beat the benchmark
systems on those highly-competitive tasks, achieving state-of-the-art
accuracies yet with substantially faster training speed.
| [
{
"version": "v1",
"created": "Sun, 23 Nov 2014 14:11:01 GMT"
},
{
"version": "v2",
"created": "Fri, 30 Jan 2015 07:41:21 GMT"
}
] | 2015-02-02T00:00:00 | [
[
"Sun",
"Xu",
""
]
] | TITLE: Structure Regularization for Structured Prediction: Theories and
Experiments
ABSTRACT: While there are many studies on weight regularization, the study on structure
regularization is rare. Many existing systems on structured prediction focus on
increasing the level of structural dependencies within the model. However, this
trend could have been misdirected, because our study suggests that complex
structures are actually harmful to generalization ability in structured
prediction. To control structure-based overfitting, we propose a structure
regularization framework via \emph{structure decomposition}, which decomposes
training samples into mini-samples with simpler structures, deriving a model
with better generalization power. We show both theoretically and empirically
that structure regularization can effectively control overfitting risk and lead
to better accuracy. As a by-product, the proposed method can also substantially
accelerate the training speed. The method and the theoretical results can apply
to general graphical models with arbitrary structures. Experiments on
well-known tasks demonstrate that our method can easily beat the benchmark
systems on those highly-competitive tasks, achieving state-of-the-art
accuracies yet with substantially faster training speed.
|
1501.05703 | Ning Zhang | Ning Zhang, Manohar Paluri, Yaniv Taigman, Rob Fergus, Lubomir Bourdev | Beyond Frontal Faces: Improving Person Recognition Using Multiple Cues | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We explore the task of recognizing peoples' identities in photo albums in an
unconstrained setting. To facilitate this, we introduce the new People In Photo
Albums (PIPA) dataset, consisting of over 60000 instances of 2000 individuals
collected from public Flickr photo albums. With only about half of the person
images containing a frontal face, the recognition task is very challenging due
to the large variations in pose, clothing, camera viewpoint, image resolution
and illumination. We propose the Pose Invariant PErson Recognition (PIPER)
method, which accumulates the cues of poselet-level person recognizers trained
by deep convolutional networks to discount for the pose variations, combined
with a face recognizer and a global recognizer. Experiments on three different
settings confirm that in our unconstrained setup PIPER significantly improves
on the performance of DeepFace, which is one of the best face recognizers as
measured on the LFW dataset.
| [
{
"version": "v1",
"created": "Fri, 23 Jan 2015 02:35:01 GMT"
},
{
"version": "v2",
"created": "Fri, 30 Jan 2015 18:48:27 GMT"
}
] | 2015-02-02T00:00:00 | [
[
"Zhang",
"Ning",
""
],
[
"Paluri",
"Manohar",
""
],
[
"Taigman",
"Yaniv",
""
],
[
"Fergus",
"Rob",
""
],
[
"Bourdev",
"Lubomir",
""
]
] | TITLE: Beyond Frontal Faces: Improving Person Recognition Using Multiple Cues
ABSTRACT: We explore the task of recognizing peoples' identities in photo albums in an
unconstrained setting. To facilitate this, we introduce the new People In Photo
Albums (PIPA) dataset, consisting of over 60000 instances of 2000 individuals
collected from public Flickr photo albums. With only about half of the person
images containing a frontal face, the recognition task is very challenging due
to the large variations in pose, clothing, camera viewpoint, image resolution
and illumination. We propose the Pose Invariant PErson Recognition (PIPER)
method, which accumulates the cues of poselet-level person recognizers trained
by deep convolutional networks to discount for the pose variations, combined
with a face recognizer and a global recognizer. Experiments on three different
settings confirm that in our unconstrained setup PIPER significantly improves
on the performance of DeepFace, which is one of the best face recognizers as
measured on the LFW dataset.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.