id
large_stringlengths 9
16
| submitter
large_stringlengths 3
64
⌀ | authors
large_stringlengths 5
20.2k
| title
large_stringlengths 5
245
| comments
large_stringlengths 1
482
⌀ | journal-ref
large_stringlengths 4
382
⌀ | doi
large_stringlengths 9
151
⌀ | report-no
large_stringlengths 2
509
⌀ | categories
large_stringlengths 5
78
| license
large_stringclasses 9
values | abstract
large_stringlengths 83
3.55k
| versions
large listlengths 1
21
| update_date
timestamp[ms]date 2007-05-23 00:00:00
2025-05-30 00:00:00
| authors_parsed
large listlengths 1
1.35k
| prompt
large_stringlengths 166
3.64k
| label
stringclasses 2
values | prob
float64 0.5
0.73
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1504.06587 | Dinesh Reddy Narapureddy | N. Dinesh Reddy, Prateek Singhal, K. Madhava Krishna | Semantic Motion Segmentation Using Dense CRF Formulation | null | null | 10.1145/2683483.2683539 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While the literature has been fairly dense in the areas of scene
understanding and semantic labeling there have been few works that make use of
motion cues to embellish semantic performance and vice versa. In this paper, we
address the problem of semantic motion segmentation, and show how semantic and
motion priors augments performance. We pro- pose an algorithm that jointly
infers the semantic class and motion labels of an object. Integrating semantic,
geometric and optical ow based constraints into a dense CRF-model we infer both
the object class as well as motion class, for each pixel. We found improvement
in performance using a fully connected CRF as compared to a standard
clique-based CRFs. For inference, we use a Mean Field approximation based
algorithm. Our method outperforms recently pro- posed motion detection
algorithms and also improves the semantic labeling compared to the
state-of-the-art Automatic Labeling Environment algorithm on the challenging
KITTI dataset especially for object classes such as pedestrians and cars that
are critical to an outdoor robotic navigation scenario.
| [
{
"version": "v1",
"created": "Fri, 24 Apr 2015 18:06:50 GMT"
}
] | 2015-04-27T00:00:00 | [
[
"Reddy",
"N. Dinesh",
""
],
[
"Singhal",
"Prateek",
""
],
[
"Krishna",
"K. Madhava",
""
]
] | TITLE: Semantic Motion Segmentation Using Dense CRF Formulation
ABSTRACT: While the literature has been fairly dense in the areas of scene
understanding and semantic labeling there have been few works that make use of
motion cues to embellish semantic performance and vice versa. In this paper, we
address the problem of semantic motion segmentation, and show how semantic and
motion priors augments performance. We pro- pose an algorithm that jointly
infers the semantic class and motion labels of an object. Integrating semantic,
geometric and optical ow based constraints into a dense CRF-model we infer both
the object class as well as motion class, for each pixel. We found improvement
in performance using a fully connected CRF as compared to a standard
clique-based CRFs. For inference, we use a Mean Field approximation based
algorithm. Our method outperforms recently pro- posed motion detection
algorithms and also improves the semantic labeling compared to the
state-of-the-art Automatic Labeling Environment algorithm on the challenging
KITTI dataset especially for object classes such as pedestrians and cars that
are critical to an outdoor robotic navigation scenario.
| no_new_dataset | 0.710377 |
1504.06591 | Konda Reddy Mopuri | Konda Reddy Mopuri and R. Venkatesh Babu | Object Level Deep Feature Pooling for Compact Image Representation | Deep Vision 2015 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Convolutional Neural Network (CNN) features have been successfully employed
in recent works as an image descriptor for various vision tasks. But the
inability of the deep CNN features to exhibit invariance to geometric
transformations and object compositions poses a great challenge for image
search. In this work, we demonstrate the effectiveness of the objectness prior
over the deep CNN features of image regions for obtaining an invariant image
representation. The proposed approach represents the image as a vector of
pooled CNN features describing the underlying objects. This representation
provides robustness to spatial layout of the objects in the scene and achieves
invariance to general geometric transformations, such as translation, rotation
and scaling. The proposed approach also leads to a compact representation of
the scene, making each image occupy a smaller memory footprint. Experiments
show that the proposed representation achieves state of the art retrieval
results on a set of challenging benchmark image datasets, while maintaining a
compact representation.
| [
{
"version": "v1",
"created": "Fri, 24 Apr 2015 18:27:25 GMT"
}
] | 2015-04-27T00:00:00 | [
[
"Mopuri",
"Konda Reddy",
""
],
[
"Babu",
"R. Venkatesh",
""
]
] | TITLE: Object Level Deep Feature Pooling for Compact Image Representation
ABSTRACT: Convolutional Neural Network (CNN) features have been successfully employed
in recent works as an image descriptor for various vision tasks. But the
inability of the deep CNN features to exhibit invariance to geometric
transformations and object compositions poses a great challenge for image
search. In this work, we demonstrate the effectiveness of the objectness prior
over the deep CNN features of image regions for obtaining an invariant image
representation. The proposed approach represents the image as a vector of
pooled CNN features describing the underlying objects. This representation
provides robustness to spatial layout of the objects in the scene and achieves
invariance to general geometric transformations, such as translation, rotation
and scaling. The proposed approach also leads to a compact representation of
the scene, making each image occupy a smaller memory footprint. Experiments
show that the proposed representation achieves state of the art retrieval
results on a set of challenging benchmark image datasets, while maintaining a
compact representation.
| no_new_dataset | 0.711017 |
1405.0425 | Manlio De Domenico | M. De Domenico, V. Nicosia, A. Arenas, V. Latora | Layer aggregation and reducibility of multilayer interconnected networks | 6 pages, 4 figures | Nature Communications 6, 6864 (2015) | 10.1038/ncomms7864 | null | physics.soc-ph cond-mat.dis-nn cs.SI physics.bio-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Many complex systems can be represented as networks composed by distinct
layers, interacting and depending on each others. For example, in biology, a
good description of the full protein-protein interactome requires, for some
organisms, up to seven distinct network layers, with thousands of
protein-protein interactions each. A fundamental open question is then how much
information is really necessary to accurately represent the structure of a
multilayer complex system, and if and when some of the layers can indeed be
aggregated. Here we introduce a method, based on information theory, to reduce
the number of layers in multilayer networks, while minimizing information loss.
We validate our approach on a set of synthetic benchmarks, and prove its
applicability to an extended data set of protein-genetic interactions, showing
cases where a strong reduction is possible and cases where it is not. Using
this method we can describe complex systems with an optimal trade--off between
accuracy and complexity.
| [
{
"version": "v1",
"created": "Fri, 2 May 2014 15:23:34 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"De Domenico",
"M.",
""
],
[
"Nicosia",
"V.",
""
],
[
"Arenas",
"A.",
""
],
[
"Latora",
"V.",
""
]
] | TITLE: Layer aggregation and reducibility of multilayer interconnected networks
ABSTRACT: Many complex systems can be represented as networks composed by distinct
layers, interacting and depending on each others. For example, in biology, a
good description of the full protein-protein interactome requires, for some
organisms, up to seven distinct network layers, with thousands of
protein-protein interactions each. A fundamental open question is then how much
information is really necessary to accurately represent the structure of a
multilayer complex system, and if and when some of the layers can indeed be
aggregated. Here we introduce a method, based on information theory, to reduce
the number of layers in multilayer networks, while minimizing information loss.
We validate our approach on a set of synthetic benchmarks, and prove its
applicability to an extended data set of protein-genetic interactions, showing
cases where a strong reduction is possible and cases where it is not. Using
this method we can describe complex systems with an optimal trade--off between
accuracy and complexity.
| no_new_dataset | 0.709831 |
1501.04587 | Naiyan Wang | Naiyan Wang, Siyi Li, Abhinav Gupta, Dit-Yan Yeung | Transferring Rich Feature Hierarchies for Robust Visual Tracking | null | null | null | null | cs.CV cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Convolutional neural network (CNN) models have demonstrated great success in
various computer vision tasks including image classification and object
detection. However, some equally important tasks such as visual tracking remain
relatively unexplored. We believe that a major hurdle that hinders the
application of CNN to visual tracking is the lack of properly labeled training
data. While existing applications that liberate the power of CNN often need an
enormous amount of training data in the order of millions, visual tracking
applications typically have only one labeled example in the first frame of each
video. We address this research issue here by pre-training a CNN offline and
then transferring the rich feature hierarchies learned to online tracking. The
CNN is also fine-tuned during online tracking to adapt to the appearance of the
tracked target specified in the first video frame. To fit the characteristics
of object tracking, we first pre-train the CNN to recognize what is an object,
and then propose to generate a probability map instead of producing a simple
class label. Using two challenging open benchmarks for performance evaluation,
our proposed tracker has demonstrated substantial improvement over other
state-of-the-art trackers.
| [
{
"version": "v1",
"created": "Mon, 19 Jan 2015 18:54:34 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Apr 2015 06:18:09 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Wang",
"Naiyan",
""
],
[
"Li",
"Siyi",
""
],
[
"Gupta",
"Abhinav",
""
],
[
"Yeung",
"Dit-Yan",
""
]
] | TITLE: Transferring Rich Feature Hierarchies for Robust Visual Tracking
ABSTRACT: Convolutional neural network (CNN) models have demonstrated great success in
various computer vision tasks including image classification and object
detection. However, some equally important tasks such as visual tracking remain
relatively unexplored. We believe that a major hurdle that hinders the
application of CNN to visual tracking is the lack of properly labeled training
data. While existing applications that liberate the power of CNN often need an
enormous amount of training data in the order of millions, visual tracking
applications typically have only one labeled example in the first frame of each
video. We address this research issue here by pre-training a CNN offline and
then transferring the rich feature hierarchies learned to online tracking. The
CNN is also fine-tuned during online tracking to adapt to the appearance of the
tracked target specified in the first video frame. To fit the characteristics
of object tracking, we first pre-train the CNN to recognize what is an object,
and then propose to generate a probability map instead of producing a simple
class label. Using two challenging open benchmarks for performance evaluation,
our proposed tracker has demonstrated substantial improvement over other
state-of-the-art trackers.
| no_new_dataset | 0.709799 |
1504.05277 | Jianxin Wu | Bin-Bin Gao and Xiu-Shen Wei and Jianxin Wu and Weiyao Lin | Deep Spatial Pyramid: The Devil is Once Again in the Details | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we show that by carefully making good choices for various
detailed but important factors in a visual recognition framework using deep
learning features, one can achieve a simple, efficient, yet highly accurate
image classification system. We first list 5 important factors, based on both
existing researches and ideas proposed in this paper. These important detailed
factors include: 1) $\ell_2$ matrix normalization is more effective than
unnormalized or $\ell_2$ vector normalization, 2) the proposed natural deep
spatial pyramid is very effective, and 3) a very small $K$ in Fisher Vectors
surprisingly achieves higher accuracy than normally used large $K$ values.
Along with other choices (convolutional activations and multiple scales), the
proposed DSP framework is not only intuitive and efficient, but also achieves
excellent classification accuracy on many benchmark datasets. For example,
DSP's accuracy on SUN397 is 59.78%, significantly higher than previous
state-of-the-art (53.86%).
| [
{
"version": "v1",
"created": "Tue, 21 Apr 2015 02:13:44 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Apr 2015 02:20:26 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Gao",
"Bin-Bin",
""
],
[
"Wei",
"Xiu-Shen",
""
],
[
"Wu",
"Jianxin",
""
],
[
"Lin",
"Weiyao",
""
]
] | TITLE: Deep Spatial Pyramid: The Devil is Once Again in the Details
ABSTRACT: In this paper we show that by carefully making good choices for various
detailed but important factors in a visual recognition framework using deep
learning features, one can achieve a simple, efficient, yet highly accurate
image classification system. We first list 5 important factors, based on both
existing researches and ideas proposed in this paper. These important detailed
factors include: 1) $\ell_2$ matrix normalization is more effective than
unnormalized or $\ell_2$ vector normalization, 2) the proposed natural deep
spatial pyramid is very effective, and 3) a very small $K$ in Fisher Vectors
surprisingly achieves higher accuracy than normally used large $K$ values.
Along with other choices (convolutional activations and multiple scales), the
proposed DSP framework is not only intuitive and efficient, but also achieves
excellent classification accuracy on many benchmark datasets. For example,
DSP's accuracy on SUN397 is 59.78%, significantly higher than previous
state-of-the-art (53.86%).
| no_new_dataset | 0.710848 |
1504.05997 | Dong Su | Dong Su, Jianneng Cao, Ninghui Li | Differentially Private Projected Histograms of Multi-Attribute Data for
Classification | null | null | null | null | cs.CR | http://creativecommons.org/licenses/by/3.0/ | In this paper, we tackle the problem of constructing a differentially private
synopsis for the classification analyses. Several the state-of-the-art methods
follow the structure of existing classification algorithms and are all
iterative, which is suboptimal due to the locally optimal choices and the
over-divided privacy budget among many sequentially composed steps. Instead, we
propose a new approach, PrivPfC, a new differentially private method for
releasing data for classification. The key idea is to privately select an
optimal partition of the underlying dataset using the given privacy budget in
one step. Given one dataset and the privacy budget, PrivPfC constructs a pool
of candidate grids where the number of cells of each grid is under a data-aware
and privacy-budget-aware threshold. After that, PrivPfC selects an optimal grid
via the exponential mechanism by using a novel quality function which minimizes
the expected number of misclassified records on which a histogram classifier is
constructed using the published grid. Finally, PrivPfC injects noise into each
cell of the selected grid and releases the noisy grid as the private synopsis
of the data. If the size of the candidate grid pool is larger than the
processing capability threshold set by the data curator, we add a step in the
beginning of PrivPfC to prune the set of attributes privately. We introduce a
modified $\chi^2$ quality function with low sensitivity and use it to evaluate
an attribute's relevance to the classification label variable. Through
extensive experiments on real datasets, we demonstrate PrivPfC's superiority
over the state-of-the-art methods.
| [
{
"version": "v1",
"created": "Wed, 22 Apr 2015 22:20:26 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Su",
"Dong",
""
],
[
"Cao",
"Jianneng",
""
],
[
"Li",
"Ninghui",
""
]
] | TITLE: Differentially Private Projected Histograms of Multi-Attribute Data for
Classification
ABSTRACT: In this paper, we tackle the problem of constructing a differentially private
synopsis for the classification analyses. Several the state-of-the-art methods
follow the structure of existing classification algorithms and are all
iterative, which is suboptimal due to the locally optimal choices and the
over-divided privacy budget among many sequentially composed steps. Instead, we
propose a new approach, PrivPfC, a new differentially private method for
releasing data for classification. The key idea is to privately select an
optimal partition of the underlying dataset using the given privacy budget in
one step. Given one dataset and the privacy budget, PrivPfC constructs a pool
of candidate grids where the number of cells of each grid is under a data-aware
and privacy-budget-aware threshold. After that, PrivPfC selects an optimal grid
via the exponential mechanism by using a novel quality function which minimizes
the expected number of misclassified records on which a histogram classifier is
constructed using the published grid. Finally, PrivPfC injects noise into each
cell of the selected grid and releases the noisy grid as the private synopsis
of the data. If the size of the candidate grid pool is larger than the
processing capability threshold set by the data curator, we add a step in the
beginning of PrivPfC to prune the set of attributes privately. We introduce a
modified $\chi^2$ quality function with low sensitivity and use it to evaluate
an attribute's relevance to the classification label variable. Through
extensive experiments on real datasets, we demonstrate PrivPfC's superiority
over the state-of-the-art methods.
| no_new_dataset | 0.709604 |
1504.05998 | Dong Su | Dong Su, Jianneng Cao, Ninghui Li, Elisa Bertino, Hongxia Jin | Differentially Private $k$-Means Clustering | null | null | null | null | cs.CR | http://creativecommons.org/licenses/by/3.0/ | There are two broad approaches for differentially private data analysis. The
interactive approach aims at developing customized differentially private
algorithms for various data mining tasks. The non-interactive approach aims at
developing differentially private algorithms that can output a synopsis of the
input dataset, which can then be used to support various data mining tasks. In
this paper we study the tradeoff of interactive vs. non-interactive approaches
and propose a hybrid approach that combines interactive and non-interactive,
using $k$-means clustering as an example. In the hybrid approach to
differentially private $k$-means clustering, one first uses a non-interactive
mechanism to publish a synopsis of the input dataset, then applies the standard
$k$-means clustering algorithm to learn $k$ cluster centroids, and finally uses
an interactive approach to further improve these cluster centroids. We analyze
the error behavior of both non-interactive and interactive approaches and use
such analysis to decide how to allocate privacy budget between the
non-interactive step and the interactive step. Results from extensive
experiments support our analysis and demonstrate the effectiveness of our
approach.
| [
{
"version": "v1",
"created": "Wed, 22 Apr 2015 22:21:30 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Su",
"Dong",
""
],
[
"Cao",
"Jianneng",
""
],
[
"Li",
"Ninghui",
""
],
[
"Bertino",
"Elisa",
""
],
[
"Jin",
"Hongxia",
""
]
] | TITLE: Differentially Private $k$-Means Clustering
ABSTRACT: There are two broad approaches for differentially private data analysis. The
interactive approach aims at developing customized differentially private
algorithms for various data mining tasks. The non-interactive approach aims at
developing differentially private algorithms that can output a synopsis of the
input dataset, which can then be used to support various data mining tasks. In
this paper we study the tradeoff of interactive vs. non-interactive approaches
and propose a hybrid approach that combines interactive and non-interactive,
using $k$-means clustering as an example. In the hybrid approach to
differentially private $k$-means clustering, one first uses a non-interactive
mechanism to publish a synopsis of the input dataset, then applies the standard
$k$-means clustering algorithm to learn $k$ cluster centroids, and finally uses
an interactive approach to further improve these cluster centroids. We analyze
the error behavior of both non-interactive and interactive approaches and use
such analysis to decide how to allocate privacy budget between the
non-interactive step and the interactive step. Results from extensive
experiments support our analysis and demonstrate the effectiveness of our
approach.
| no_new_dataset | 0.710038 |
1504.06055 | Naiyan Wang | Naiyan Wang, Jianping Shi, Dit-Yan Yeung, Jiaya Jia | Understanding and Diagnosing Visual Tracking Systems | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Several benchmark datasets for visual tracking research have been proposed in
recent years. Despite their usefulness, whether they are sufficient for
understanding and diagnosing the strengths and weaknesses of different trackers
remains questionable. To address this issue, we propose a framework by breaking
a tracker down into five constituent parts, namely, motion model, feature
extractor, observation model, model updater, and ensemble post-processor. We
then conduct ablative experiments on each component to study how it affects the
overall result. Surprisingly, our findings are discrepant with some common
beliefs in the visual tracking research community. We find that the feature
extractor plays the most important role in a tracker. On the other hand,
although the observation model is the focus of many studies, we find that it
often brings no significant improvement. Moreover, the motion model and model
updater contain many details that could affect the result. Also, the ensemble
post-processor can improve the result substantially when the constituent
trackers have high diversity. Based on our findings, we put together some very
elementary building blocks to give a basic tracker which is competitive in
performance to the state-of-the-art trackers. We believe our framework can
provide a solid baseline when conducting controlled experiments for visual
tracking research.
| [
{
"version": "v1",
"created": "Thu, 23 Apr 2015 06:37:29 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Wang",
"Naiyan",
""
],
[
"Shi",
"Jianping",
""
],
[
"Yeung",
"Dit-Yan",
""
],
[
"Jia",
"Jiaya",
""
]
] | TITLE: Understanding and Diagnosing Visual Tracking Systems
ABSTRACT: Several benchmark datasets for visual tracking research have been proposed in
recent years. Despite their usefulness, whether they are sufficient for
understanding and diagnosing the strengths and weaknesses of different trackers
remains questionable. To address this issue, we propose a framework by breaking
a tracker down into five constituent parts, namely, motion model, feature
extractor, observation model, model updater, and ensemble post-processor. We
then conduct ablative experiments on each component to study how it affects the
overall result. Surprisingly, our findings are discrepant with some common
beliefs in the visual tracking research community. We find that the feature
extractor plays the most important role in a tracker. On the other hand,
although the observation model is the focus of many studies, we find that it
often brings no significant improvement. Moreover, the motion model and model
updater contain many details that could affect the result. Also, the ensemble
post-processor can improve the result substantially when the constituent
trackers have high diversity. Based on our findings, we put together some very
elementary building blocks to give a basic tracker which is competitive in
performance to the state-of-the-art trackers. We believe our framework can
provide a solid baseline when conducting controlled experiments for visual
tracking research.
| no_new_dataset | 0.708389 |
1504.06078 | Nicolas Turenne | Nicolas Turenne, Tien Phan | x.ent: R Package for Entities and Relations Extraction based on
Unsupervised Learning and Document Structure | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Relation extraction with accurate precision is still a challenge when
processing full text databases. We propose an approach based on cooccurrence
analysis in each document for which we used document organization to improve
accuracy of relation extraction. This approach is implemented in a R package
called \emph{x.ent}. Another facet of extraction relies on use of extracted
relation into a querying system for expert end-users. Two datasets had been
used. One of them gets interest from specialists of epidemiology in plant
health. For this dataset usage is dedicated to plant-disease exploration
through agricultural information news. An open-data platform exploits exports
from \emph{x.ent} and is publicly available.
| [
{
"version": "v1",
"created": "Thu, 23 Apr 2015 08:28:01 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Turenne",
"Nicolas",
""
],
[
"Phan",
"Tien",
""
]
] | TITLE: x.ent: R Package for Entities and Relations Extraction based on
Unsupervised Learning and Document Structure
ABSTRACT: Relation extraction with accurate precision is still a challenge when
processing full text databases. We propose an approach based on cooccurrence
analysis in each document for which we used document organization to improve
accuracy of relation extraction. This approach is implemented in a R package
called \emph{x.ent}. Another facet of extraction relies on use of extracted
relation into a querying system for expert end-users. Two datasets had been
used. One of them gets interest from specialists of epidemiology in plant
health. For this dataset usage is dedicated to plant-disease exploration
through agricultural information news. An open-data platform exploits exports
from \emph{x.ent} and is publicly available.
| no_new_dataset | 0.707222 |
1504.06133 | Anguelos Nicolaou | Anguelos Nicolaou, Andrew D. Bagdanov, Marcus Liwicki, Dimosthenis
Karatzas | Sparse Radial Sampling LBP for Writer Identification | Submitted to the 13th International Conference on Document Analysis
and Recognition (ICDAR 2015) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we present the use of Sparse Radial Sampling Local Binary
Patterns, a variant of Local Binary Patterns (LBP) for text-as-texture
classification. By adapting and extending the standard LBP operator to the
particularities of text we get a generic text-as-texture classification scheme
and apply it to writer identification. In experiments on CVL and ICDAR 2013
datasets, the proposed feature-set demonstrates State-Of-the-Art (SOA)
performance. Among the SOA, the proposed method is the only one that is based
on dense extraction of a single local feature descriptor. This makes it fast
and applicable at the earliest stages in a DIA pipeline without the need for
segmentation, binarization, or extraction of multiple features.
| [
{
"version": "v1",
"created": "Thu, 23 Apr 2015 11:51:53 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Nicolaou",
"Anguelos",
""
],
[
"Bagdanov",
"Andrew D.",
""
],
[
"Liwicki",
"Marcus",
""
],
[
"Karatzas",
"Dimosthenis",
""
]
] | TITLE: Sparse Radial Sampling LBP for Writer Identification
ABSTRACT: In this paper we present the use of Sparse Radial Sampling Local Binary
Patterns, a variant of Local Binary Patterns (LBP) for text-as-texture
classification. By adapting and extending the standard LBP operator to the
particularities of text we get a generic text-as-texture classification scheme
and apply it to writer identification. In experiments on CVL and ICDAR 2013
datasets, the proposed feature-set demonstrates State-Of-the-Art (SOA)
performance. Among the SOA, the proposed method is the only one that is based
on dense extraction of a single local feature descriptor. This makes it fast
and applicable at the earliest stages in a DIA pipeline without the need for
segmentation, binarization, or extraction of multiple features.
| no_new_dataset | 0.712226 |
1504.06151 | Nauman Shahid | Nauman Shahid, Vassilis Kalofolias, Xavier Bresson, Michael Bronstein
and Pierre Vandergheynst | Robust Principal Component Analysis on Graphs | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Principal Component Analysis (PCA) is the most widely used tool for linear
dimensionality reduction and clustering. Still it is highly sensitive to
outliers and does not scale well with respect to the number of data samples.
Robust PCA solves the first issue with a sparse penalty term. The second issue
can be handled with the matrix factorization model, which is however
non-convex. Besides, PCA based clustering can also be enhanced by using a graph
of data similarity. In this article, we introduce a new model called "Robust
PCA on Graphs" which incorporates spectral graph regularization into the Robust
PCA framework. Our proposed model benefits from 1) the robustness of principal
components to occlusions and missing values, 2) enhanced low-rank recovery, 3)
improved clustering property due to the graph smoothness assumption on the
low-rank matrix, and 4) convexity of the resulting optimization problem.
Extensive experiments on 8 benchmark, 3 video and 2 artificial datasets with
corruptions clearly reveal that our model outperforms 10 other state-of-the-art
models in its clustering and low-rank recovery tasks.
| [
{
"version": "v1",
"created": "Thu, 23 Apr 2015 12:39:40 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Shahid",
"Nauman",
""
],
[
"Kalofolias",
"Vassilis",
""
],
[
"Bresson",
"Xavier",
""
],
[
"Bronstein",
"Michael",
""
],
[
"Vandergheynst",
"Pierre",
""
]
] | TITLE: Robust Principal Component Analysis on Graphs
ABSTRACT: Principal Component Analysis (PCA) is the most widely used tool for linear
dimensionality reduction and clustering. Still it is highly sensitive to
outliers and does not scale well with respect to the number of data samples.
Robust PCA solves the first issue with a sparse penalty term. The second issue
can be handled with the matrix factorization model, which is however
non-convex. Besides, PCA based clustering can also be enhanced by using a graph
of data similarity. In this article, we introduce a new model called "Robust
PCA on Graphs" which incorporates spectral graph regularization into the Robust
PCA framework. Our proposed model benefits from 1) the robustness of principal
components to occlusions and missing values, 2) enhanced low-rank recovery, 3)
improved clustering property due to the graph smoothness assumption on the
low-rank matrix, and 4) convexity of the resulting optimization problem.
Extensive experiments on 8 benchmark, 3 video and 2 artificial datasets with
corruptions clearly reveal that our model outperforms 10 other state-of-the-art
models in its clustering and low-rank recovery tasks.
| no_new_dataset | 0.711425 |
1504.06165 | Nitish Gupta | Nitish Gupta, Sameer Singh | Collectively Embedding Multi-Relational Data for Predicting User
Preferences | 10 pages, 5 figures | null | null | null | cs.LG cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Matrix factorization has found incredible success and widespread application
as a collaborative filtering based approach to recommendations. Unfortunately,
incorporating additional sources of evidence, especially ones that are
incomplete and noisy, is quite difficult to achieve in such models, however, is
often crucial for obtaining further gains in accuracy. For example, additional
information about businesses from reviews, categories, and attributes should be
leveraged for predicting user preferences, even though this information is
often inaccurate and partially-observed. Instead of creating customized methods
that are specific to each type of evidences, in this paper we present a generic
approach to factorization of relational data that collectively models all the
relations in the database. By learning a set of embeddings that are shared
across all the relations, the model is able to incorporate observed information
from all the relations, while also predicting all the relations of interest.
Our evaluation on multiple Amazon and Yelp datasets demonstrates effective
utilization of additional information for held-out preference prediction, but
further, we present accurate models even for the cold-starting businesses and
products for which we do not observe any ratings or reviews. We also illustrate
the capability of the model in imputing missing information and jointly
visualizing words, categories, and attribute factors.
| [
{
"version": "v1",
"created": "Thu, 23 Apr 2015 13:07:24 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Gupta",
"Nitish",
""
],
[
"Singh",
"Sameer",
""
]
] | TITLE: Collectively Embedding Multi-Relational Data for Predicting User
Preferences
ABSTRACT: Matrix factorization has found incredible success and widespread application
as a collaborative filtering based approach to recommendations. Unfortunately,
incorporating additional sources of evidence, especially ones that are
incomplete and noisy, is quite difficult to achieve in such models, however, is
often crucial for obtaining further gains in accuracy. For example, additional
information about businesses from reviews, categories, and attributes should be
leveraged for predicting user preferences, even though this information is
often inaccurate and partially-observed. Instead of creating customized methods
that are specific to each type of evidences, in this paper we present a generic
approach to factorization of relational data that collectively models all the
relations in the database. By learning a set of embeddings that are shared
across all the relations, the model is able to incorporate observed information
from all the relations, while also predicting all the relations of interest.
Our evaluation on multiple Amazon and Yelp datasets demonstrates effective
utilization of additional information for held-out preference prediction, but
further, we present accurate models even for the cold-starting businesses and
products for which we do not observe any ratings or reviews. We also illustrate
the capability of the model in imputing missing information and jointly
visualizing words, categories, and attribute factors.
| no_new_dataset | 0.709422 |
1504.06266 | Hamid Tizhoosh | Ahmed Othman, Hamid R. Tizhoosh, Farzad Khalvati | Evolving Fuzzy Image Segmentation with Self-Configuration | Benchmark data (35 breast ultrasound images with gold standard
segments) available; 11 pages, 4 algorithms, 6 figures, 5 tables; | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/3.0/ | Current image segmentation techniques usually require that the user tune
several parameters in order to obtain maximum segmentation accuracy, a
computationally inefficient approach, especially when a large number of images
must be processed sequentially in daily practice. The use of evolving fuzzy
systems for designing a method that automatically adjusts parameters to segment
medical images according to the quality expectation of expert users has been
proposed recently (Evolving fuzzy image segmentation EFIS). However, EFIS
suffers from a few limitations when used in practice mainly due to some fixed
parameters. For instance, EFIS depends on auto-detection of the object of
interest for feature calculation, a task that is highly application-dependent.
This shortcoming limits the applicability of EFIS, which was proposed with the
ultimate goal of offering a generic but adjustable segmentation scheme. In this
paper, a new version of EFIS is proposed to overcome these limitations. The new
EFIS, called self-configuring EFIS (SC-EFIS), uses available training data to
self-estimate the parameters that are fixed in EFIS. As well, the proposed
SC-EFIS relies on a feature selection process that does not require
auto-detection of an ROI. The proposed SC-EFIS was evaluated using the same
segmentation algorithms and the same dataset as for EFIS. The results show that
SC-EFIS can provide the same results as EFIS but with a higher level of
automation.
| [
{
"version": "v1",
"created": "Thu, 23 Apr 2015 17:23:09 GMT"
}
] | 2015-04-24T00:00:00 | [
[
"Othman",
"Ahmed",
""
],
[
"Tizhoosh",
"Hamid R.",
""
],
[
"Khalvati",
"Farzad",
""
]
] | TITLE: Evolving Fuzzy Image Segmentation with Self-Configuration
ABSTRACT: Current image segmentation techniques usually require that the user tune
several parameters in order to obtain maximum segmentation accuracy, a
computationally inefficient approach, especially when a large number of images
must be processed sequentially in daily practice. The use of evolving fuzzy
systems for designing a method that automatically adjusts parameters to segment
medical images according to the quality expectation of expert users has been
proposed recently (Evolving fuzzy image segmentation EFIS). However, EFIS
suffers from a few limitations when used in practice mainly due to some fixed
parameters. For instance, EFIS depends on auto-detection of the object of
interest for feature calculation, a task that is highly application-dependent.
This shortcoming limits the applicability of EFIS, which was proposed with the
ultimate goal of offering a generic but adjustable segmentation scheme. In this
paper, a new version of EFIS is proposed to overcome these limitations. The new
EFIS, called self-configuring EFIS (SC-EFIS), uses available training data to
self-estimate the parameters that are fixed in EFIS. As well, the proposed
SC-EFIS relies on a feature selection process that does not require
auto-detection of an ROI. The proposed SC-EFIS was evaluated using the same
segmentation algorithms and the same dataset as for EFIS. The results show that
SC-EFIS can provide the same results as EFIS but with a higher level of
automation.
| no_new_dataset | 0.711243 |
1504.05451 | Jing Yang | Qingshan Liu, Jing Yang, Kaihua Zhang, Yi Wu | Adaptive Compressive Tracking via Online Vector Boosting Feature
Selection | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, the compressive tracking (CT) method has attracted much attention
due to its high efficiency, but it cannot well deal with the large scale target
appearance variations due to its data-independent random projection matrix that
results in less discriminative features. To address this issue, in this paper
we propose an adaptive CT approach, which selects the most discriminative
features to design an effective appearance model. Our method significantly
improves CT in three aspects: Firstly, the most discriminative features are
selected via an online vector boosting method. Secondly, the object
representation is updated in an effective online manner, which preserves the
stable features while filtering out the noisy ones. Finally, a simple and
effective trajectory rectification approach is adopted that can make the
estimated location more accurate. Extensive experiments on the CVPR2013
tracking benchmark demonstrate the superior performance of our algorithm
compared over state-of-the-art tracking algorithms.
| [
{
"version": "v1",
"created": "Tue, 21 Apr 2015 14:55:07 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Apr 2015 01:27:08 GMT"
}
] | 2015-04-23T00:00:00 | [
[
"Liu",
"Qingshan",
""
],
[
"Yang",
"Jing",
""
],
[
"Zhang",
"Kaihua",
""
],
[
"Wu",
"Yi",
""
]
] | TITLE: Adaptive Compressive Tracking via Online Vector Boosting Feature
Selection
ABSTRACT: Recently, the compressive tracking (CT) method has attracted much attention
due to its high efficiency, but it cannot well deal with the large scale target
appearance variations due to its data-independent random projection matrix that
results in less discriminative features. To address this issue, in this paper
we propose an adaptive CT approach, which selects the most discriminative
features to design an effective appearance model. Our method significantly
improves CT in three aspects: Firstly, the most discriminative features are
selected via an online vector boosting method. Secondly, the object
representation is updated in an effective online manner, which preserves the
stable features while filtering out the noisy ones. Finally, a simple and
effective trajectory rectification approach is adopted that can make the
estimated location more accurate. Extensive experiments on the CVPR2013
tracking benchmark demonstrate the superior performance of our algorithm
compared over state-of-the-art tracking algorithms.
| no_new_dataset | 0.710409 |
1504.05766 | Muharrem D\"u\u{g}enci | Muharrem D\"u\u{g}enci | Honeybees-inspired heuristic algorithms for numerical optimisation | 17 pages, 3 Figures, 6 Tables | null | null | KU-IE-MD-001 | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Swarm intelligence is all about developing collective behaviours to solve
complex, ill-structured and large-scale problems. Efficiency in collective
behaviours depends on how to harmonise the individual contributions so that a
complementary collective effort can be achieved to offer a useful solution. The
main points in organising the harmony remains as managing the diversification
and intensification actions appropriately, where the efficiency of collective
behaviours depends on blending these two actions appropriately. In this study,
two swarm intelligence algorithms inspired of natural honeybee colonies have
been overviewed with many respects and two new revisions and a hybrid version
have been studied to improve the efficiencies in solving numerical optimisation
problems, which are well-known hard benchmarks. Consequently, the revisions and
especially the hybrid algorithm proposed have outperformed the two original bee
algorithms in solving these very hard numerical optimisation benchmarks.
| [
{
"version": "v1",
"created": "Wed, 22 Apr 2015 12:46:07 GMT"
}
] | 2015-04-23T00:00:00 | [
[
"Düğenci",
"Muharrem",
""
]
] | TITLE: Honeybees-inspired heuristic algorithms for numerical optimisation
ABSTRACT: Swarm intelligence is all about developing collective behaviours to solve
complex, ill-structured and large-scale problems. Efficiency in collective
behaviours depends on how to harmonise the individual contributions so that a
complementary collective effort can be achieved to offer a useful solution. The
main points in organising the harmony remains as managing the diversification
and intensification actions appropriately, where the efficiency of collective
behaviours depends on blending these two actions appropriately. In this study,
two swarm intelligence algorithms inspired of natural honeybee colonies have
been overviewed with many respects and two new revisions and a hybrid version
have been studied to improve the efficiencies in solving numerical optimisation
problems, which are well-known hard benchmarks. Consequently, the revisions and
especially the hybrid algorithm proposed have outperformed the two original bee
algorithms in solving these very hard numerical optimisation benchmarks.
| no_new_dataset | 0.708427 |
1504.05809 | Xianbiao Qi | Xianbiao Qi, Guoying Zhao, Linlin Shen, Qingquan Li, Matti Pietikainen | LOAD: Local Orientation Adaptive Descriptor for Texture and Material
Classification | 13 pages, 7 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose a novel local feature, called Local Orientation
Adaptive Descriptor (LOAD), to capture regional texture in an image. In LOAD,
we proposed to define point description on an Adaptive Coordinate System (ACS),
adopt a binary sequence descriptor to capture relationships between one point
and its neighbors and use multi-scale strategy to enhance the discriminative
power of the descriptor. The proposed LOAD enjoys not only discriminative power
to capture the texture information, but also has strong robustness to
illumination variation and image rotation. Extensive experiments on benchmark
data sets of texture classification and real-world material recognition show
that the proposed LOAD yields the state-of-the-art performance. It is worth to
mention that we achieve a 65.4\% classification accuracy-- which is, to the
best of our knowledge, the highest record by far --on Flickr Material Database
by using a single feature. Moreover, by combining LOAD with the feature
extracted by Convolutional Neural Networks (CNN), we obtain significantly
better performance than both the LOAD and CNN. This result confirms that the
LOAD is complementary to the learning-based features.
| [
{
"version": "v1",
"created": "Wed, 22 Apr 2015 13:59:49 GMT"
}
] | 2015-04-23T00:00:00 | [
[
"Qi",
"Xianbiao",
""
],
[
"Zhao",
"Guoying",
""
],
[
"Shen",
"Linlin",
""
],
[
"Li",
"Qingquan",
""
],
[
"Pietikainen",
"Matti",
""
]
] | TITLE: LOAD: Local Orientation Adaptive Descriptor for Texture and Material
Classification
ABSTRACT: In this paper, we propose a novel local feature, called Local Orientation
Adaptive Descriptor (LOAD), to capture regional texture in an image. In LOAD,
we proposed to define point description on an Adaptive Coordinate System (ACS),
adopt a binary sequence descriptor to capture relationships between one point
and its neighbors and use multi-scale strategy to enhance the discriminative
power of the descriptor. The proposed LOAD enjoys not only discriminative power
to capture the texture information, but also has strong robustness to
illumination variation and image rotation. Extensive experiments on benchmark
data sets of texture classification and real-world material recognition show
that the proposed LOAD yields the state-of-the-art performance. It is worth to
mention that we achieve a 65.4\% classification accuracy-- which is, to the
best of our knowledge, the highest record by far --on Flickr Material Database
by using a single feature. Moreover, by combining LOAD with the feature
extracted by Convolutional Neural Networks (CNN), we obtain significantly
better performance than both the LOAD and CNN. This result confirms that the
LOAD is complementary to the learning-based features.
| no_new_dataset | 0.711825 |
1504.05880 | Shiva Kasiviswanathan | Shiva Prasad Kasiviswanathan and Mark Rudelson | Spectral Norm of Random Kernel Matrices with Applications to Privacy | 16 pages, 1 Figure | null | null | null | stat.ML cs.CR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Kernel methods are an extremely popular set of techniques used for many
important machine learning and data analysis applications. In addition to
having good practical performances, these methods are supported by a
well-developed theory. Kernel methods use an implicit mapping of the input data
into a high dimensional feature space defined by a kernel function, i.e., a
function returning the inner product between the images of two data points in
the feature space. Central to any kernel method is the kernel matrix, which is
built by evaluating the kernel function on a given sample dataset.
In this paper, we initiate the study of non-asymptotic spectral theory of
random kernel matrices. These are n x n random matrices whose (i,j)th entry is
obtained by evaluating the kernel function on $x_i$ and $x_j$, where
$x_1,...,x_n$ are a set of n independent random high-dimensional vectors. Our
main contribution is to obtain tight upper bounds on the spectral norm (largest
eigenvalue) of random kernel matrices constructed by commonly used kernel
functions based on polynomials and Gaussian radial basis.
As an application of these results, we provide lower bounds on the distortion
needed for releasing the coefficients of kernel ridge regression under
attribute privacy, a general privacy notion which captures a large class of
privacy definitions. Kernel ridge regression is standard method for performing
non-parametric regression that regularly outperforms traditional regression
approaches in various domains. Our privacy distortion lower bounds are the
first for any kernel technique, and our analysis assumes realistic scenarios
for the input, unlike all previous lower bounds for other release problems
which only hold under very restrictive input settings.
| [
{
"version": "v1",
"created": "Wed, 22 Apr 2015 16:54:48 GMT"
}
] | 2015-04-23T00:00:00 | [
[
"Kasiviswanathan",
"Shiva Prasad",
""
],
[
"Rudelson",
"Mark",
""
]
] | TITLE: Spectral Norm of Random Kernel Matrices with Applications to Privacy
ABSTRACT: Kernel methods are an extremely popular set of techniques used for many
important machine learning and data analysis applications. In addition to
having good practical performances, these methods are supported by a
well-developed theory. Kernel methods use an implicit mapping of the input data
into a high dimensional feature space defined by a kernel function, i.e., a
function returning the inner product between the images of two data points in
the feature space. Central to any kernel method is the kernel matrix, which is
built by evaluating the kernel function on a given sample dataset.
In this paper, we initiate the study of non-asymptotic spectral theory of
random kernel matrices. These are n x n random matrices whose (i,j)th entry is
obtained by evaluating the kernel function on $x_i$ and $x_j$, where
$x_1,...,x_n$ are a set of n independent random high-dimensional vectors. Our
main contribution is to obtain tight upper bounds on the spectral norm (largest
eigenvalue) of random kernel matrices constructed by commonly used kernel
functions based on polynomials and Gaussian radial basis.
As an application of these results, we provide lower bounds on the distortion
needed for releasing the coefficients of kernel ridge regression under
attribute privacy, a general privacy notion which captures a large class of
privacy definitions. Kernel ridge regression is standard method for performing
non-parametric regression that regularly outperforms traditional regression
approaches in various domains. Our privacy distortion lower bounds are the
first for any kernel technique, and our analysis assumes realistic scenarios
for the input, unlike all previous lower bounds for other release problems
which only hold under very restrictive input settings.
| no_new_dataset | 0.709617 |
1411.4555 | Samy Bengio | Oriol Vinyals, Alexander Toshev, Samy Bengio, Dumitru Erhan | Show and Tell: A Neural Image Caption Generator | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Automatically describing the content of an image is a fundamental problem in
artificial intelligence that connects computer vision and natural language
processing. In this paper, we present a generative model based on a deep
recurrent architecture that combines recent advances in computer vision and
machine translation and that can be used to generate natural sentences
describing an image. The model is trained to maximize the likelihood of the
target description sentence given the training image. Experiments on several
datasets show the accuracy of the model and the fluency of the language it
learns solely from image descriptions. Our model is often quite accurate, which
we verify both qualitatively and quantitatively. For instance, while the
current state-of-the-art BLEU-1 score (the higher the better) on the Pascal
dataset is 25, our approach yields 59, to be compared to human performance
around 69. We also show BLEU-1 score improvements on Flickr30k, from 56 to 66,
and on SBU, from 19 to 28. Lastly, on the newly released COCO dataset, we
achieve a BLEU-4 of 27.7, which is the current state-of-the-art.
| [
{
"version": "v1",
"created": "Mon, 17 Nov 2014 17:15:41 GMT"
},
{
"version": "v2",
"created": "Mon, 20 Apr 2015 22:26:11 GMT"
}
] | 2015-04-22T00:00:00 | [
[
"Vinyals",
"Oriol",
""
],
[
"Toshev",
"Alexander",
""
],
[
"Bengio",
"Samy",
""
],
[
"Erhan",
"Dumitru",
""
]
] | TITLE: Show and Tell: A Neural Image Caption Generator
ABSTRACT: Automatically describing the content of an image is a fundamental problem in
artificial intelligence that connects computer vision and natural language
processing. In this paper, we present a generative model based on a deep
recurrent architecture that combines recent advances in computer vision and
machine translation and that can be used to generate natural sentences
describing an image. The model is trained to maximize the likelihood of the
target description sentence given the training image. Experiments on several
datasets show the accuracy of the model and the fluency of the language it
learns solely from image descriptions. Our model is often quite accurate, which
we verify both qualitatively and quantitatively. For instance, while the
current state-of-the-art BLEU-1 score (the higher the better) on the Pascal
dataset is 25, our approach yields 59, to be compared to human performance
around 69. We also show BLEU-1 score improvements on Flickr30k, from 56 to 66,
and on SBU, from 19 to 28. Lastly, on the newly released COCO dataset, we
achieve a BLEU-4 of 27.7, which is the current state-of-the-art.
| no_new_dataset | 0.705354 |
1412.4480 | Long Zheng | Long Zheng, Xiaofei Liao, Bingsheng He, Song Wu, Hai Jin | On Performance Debugging of Unnecessary Lock Contentions on Multicore
Processors: A Replay-based Approach | 18 pages, 19 figures, 3 tables | null | null | null | cs.PL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Locks have been widely used as an effective synchronization mechanism among
processes and threads. However, we observe that a large number of false
inter-thread dependencies (i.e., unnecessary lock contentions) exist during the
program execution on multicore processors, thereby incurring significant
performance overhead. This paper presents a performance debugging framework,
PERFPLAY, to facilitate a comprehensive and in-depth understanding of the
performance impact of unnecessary lock contentions. The core technique of our
debugging framework is trace replay. Specifically, PERFPLAY records the program
execution trace, on the basis of which the unnecessary lock contentions can be
identified through trace analysis. We then propose a novel technique of trace
transformation to transform these identified unnecessary lock contentions in
the original trace into the correct pattern as a new trace free of unnecessary
lock contentions. Through replaying both traces, PERFPLAY can quantify the
performance impact of unnecessary lock contentions. To demonstrate the
effectiveness of our debugging framework, we study five real-world programs and
PARSEC benchmarks. Our experimental results demonstrate the significant
performance overhead of unnecessary lock contentions, and the effectiveness of
PERFPLAY in identifying the performance critical unnecessary lock contentions
in real applications.
| [
{
"version": "v1",
"created": "Mon, 15 Dec 2014 07:22:40 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Apr 2015 09:28:21 GMT"
}
] | 2015-04-22T00:00:00 | [
[
"Zheng",
"Long",
""
],
[
"Liao",
"Xiaofei",
""
],
[
"He",
"Bingsheng",
""
],
[
"Wu",
"Song",
""
],
[
"Jin",
"Hai",
""
]
] | TITLE: On Performance Debugging of Unnecessary Lock Contentions on Multicore
Processors: A Replay-based Approach
ABSTRACT: Locks have been widely used as an effective synchronization mechanism among
processes and threads. However, we observe that a large number of false
inter-thread dependencies (i.e., unnecessary lock contentions) exist during the
program execution on multicore processors, thereby incurring significant
performance overhead. This paper presents a performance debugging framework,
PERFPLAY, to facilitate a comprehensive and in-depth understanding of the
performance impact of unnecessary lock contentions. The core technique of our
debugging framework is trace replay. Specifically, PERFPLAY records the program
execution trace, on the basis of which the unnecessary lock contentions can be
identified through trace analysis. We then propose a novel technique of trace
transformation to transform these identified unnecessary lock contentions in
the original trace into the correct pattern as a new trace free of unnecessary
lock contentions. Through replaying both traces, PERFPLAY can quantify the
performance impact of unnecessary lock contentions. To demonstrate the
effectiveness of our debugging framework, we study five real-world programs and
PARSEC benchmarks. Our experimental results demonstrate the significant
performance overhead of unnecessary lock contentions, and the effectiveness of
PERFPLAY in identifying the performance critical unnecessary lock contentions
in real applications.
| no_new_dataset | 0.705011 |
1412.6830 | Forest Agostinelli | Forest Agostinelli, Matthew Hoffman, Peter Sadowski, Pierre Baldi | Learning Activation Functions to Improve Deep Neural Networks | Accepted as a workshop paper contribution at the International
Conference on Learning Representations (ICLR) 2015 | null | null | null | cs.NE cs.CV cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Artificial neural networks typically have a fixed, non-linear activation
function at each neuron. We have designed a novel form of piecewise linear
activation function that is learned independently for each neuron using
gradient descent. With this adaptive activation function, we are able to
improve upon deep neural network architectures composed of static rectified
linear units, achieving state-of-the-art performance on CIFAR-10 (7.51%),
CIFAR-100 (30.83%), and a benchmark from high-energy physics involving Higgs
boson decay modes.
| [
{
"version": "v1",
"created": "Sun, 21 Dec 2014 20:20:21 GMT"
},
{
"version": "v2",
"created": "Fri, 27 Feb 2015 21:44:41 GMT"
},
{
"version": "v3",
"created": "Tue, 21 Apr 2015 08:05:02 GMT"
}
] | 2015-04-22T00:00:00 | [
[
"Agostinelli",
"Forest",
""
],
[
"Hoffman",
"Matthew",
""
],
[
"Sadowski",
"Peter",
""
],
[
"Baldi",
"Pierre",
""
]
] | TITLE: Learning Activation Functions to Improve Deep Neural Networks
ABSTRACT: Artificial neural networks typically have a fixed, non-linear activation
function at each neuron. We have designed a novel form of piecewise linear
activation function that is learned independently for each neuron using
gradient descent. With this adaptive activation function, we are able to
improve upon deep neural network architectures composed of static rectified
linear units, achieving state-of-the-art performance on CIFAR-10 (7.51%),
CIFAR-100 (30.83%), and a benchmark from high-energy physics involving Higgs
boson decay modes.
| no_new_dataset | 0.710025 |
1502.02710 | Nikos Karampatziakis | Nikos Karampatziakis, Paul Mineiro | Scalable Multilabel Prediction via Randomized Methods | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Modeling the dependence between outputs is a fundamental challenge in
multilabel classification. In this work we show that a generic regularized
nonlinearity mapping independent predictions to joint predictions is sufficient
to achieve state-of-the-art performance on a variety of benchmark problems.
Crucially, we compute the joint predictions without ever obtaining any
independent predictions, while incorporating low-rank and smoothness
regularization. We achieve this by leveraging randomized algorithms for matrix
decomposition and kernel approximation. Furthermore, our techniques are
applicable to the multiclass setting. We apply our method to a variety of
multiclass and multilabel data sets, obtaining state-of-the-art results.
| [
{
"version": "v1",
"created": "Mon, 9 Feb 2015 22:18:26 GMT"
},
{
"version": "v2",
"created": "Mon, 20 Apr 2015 21:08:19 GMT"
}
] | 2015-04-22T00:00:00 | [
[
"Karampatziakis",
"Nikos",
""
],
[
"Mineiro",
"Paul",
""
]
] | TITLE: Scalable Multilabel Prediction via Randomized Methods
ABSTRACT: Modeling the dependence between outputs is a fundamental challenge in
multilabel classification. In this work we show that a generic regularized
nonlinearity mapping independent predictions to joint predictions is sufficient
to achieve state-of-the-art performance on a variety of benchmark problems.
Crucially, we compute the joint predictions without ever obtaining any
independent predictions, while incorporating low-rank and smoothness
regularization. We achieve this by leveraging randomized algorithms for matrix
decomposition and kernel approximation. Furthermore, our techniques are
applicable to the multiclass setting. We apply our method to a variety of
multiclass and multilabel data sets, obtaining state-of-the-art results.
| no_new_dataset | 0.710415 |
1502.02766 | Sachin Sudhakar Farfade | Sachin Sudhakar Farfade, Mohammad Saberian, Li-Jia Li | Multi-view Face Detection Using Deep Convolutional Neural Networks | in International Conference on Multimedia Retrieval 2015 (ICMR) | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/3.0/ | In this paper we consider the problem of multi-view face detection. While
there has been significant research on this problem, current state-of-the-art
approaches for this task require annotation of facial landmarks, e.g. TSM [25],
or annotation of face poses [28, 22]. They also require training dozens of
models to fully capture faces in all orientations, e.g. 22 models in HeadHunter
method [22]. In this paper we propose Deep Dense Face Detector (DDFD), a method
that does not require pose/landmark annotation and is able to detect faces in a
wide range of orientations using a single model based on deep convolutional
neural networks. The proposed method has minimal complexity; unlike other
recent deep learning object detection methods [9], it does not require
additional components such as segmentation, bounding-box regression, or SVM
classifiers. Furthermore, we analyzed scores of the proposed face detector for
faces in different orientations and found that 1) the proposed method is able
to detect faces from different angles and can handle occlusion to some extent,
2) there seems to be a correlation between dis- tribution of positive examples
in the training set and scores of the proposed face detector. The latter
suggests that the proposed methods performance can be further improved by using
better sampling strategies and more sophisticated data augmentation techniques.
Evaluations on popular face detection benchmark datasets show that our
single-model face detector algorithm has similar or better performance compared
to the previous methods, which are more complex and require annotations of
either different poses or facial landmarks.
| [
{
"version": "v1",
"created": "Tue, 10 Feb 2015 03:15:21 GMT"
},
{
"version": "v2",
"created": "Wed, 4 Mar 2015 10:07:20 GMT"
},
{
"version": "v3",
"created": "Mon, 20 Apr 2015 20:18:57 GMT"
}
] | 2015-04-22T00:00:00 | [
[
"Farfade",
"Sachin Sudhakar",
""
],
[
"Saberian",
"Mohammad",
""
],
[
"Li",
"Li-Jia",
""
]
] | TITLE: Multi-view Face Detection Using Deep Convolutional Neural Networks
ABSTRACT: In this paper we consider the problem of multi-view face detection. While
there has been significant research on this problem, current state-of-the-art
approaches for this task require annotation of facial landmarks, e.g. TSM [25],
or annotation of face poses [28, 22]. They also require training dozens of
models to fully capture faces in all orientations, e.g. 22 models in HeadHunter
method [22]. In this paper we propose Deep Dense Face Detector (DDFD), a method
that does not require pose/landmark annotation and is able to detect faces in a
wide range of orientations using a single model based on deep convolutional
neural networks. The proposed method has minimal complexity; unlike other
recent deep learning object detection methods [9], it does not require
additional components such as segmentation, bounding-box regression, or SVM
classifiers. Furthermore, we analyzed scores of the proposed face detector for
faces in different orientations and found that 1) the proposed method is able
to detect faces from different angles and can handle occlusion to some extent,
2) there seems to be a correlation between dis- tribution of positive examples
in the training set and scores of the proposed face detector. The latter
suggests that the proposed methods performance can be further improved by using
better sampling strategies and more sophisticated data augmentation techniques.
Evaluations on popular face detection benchmark datasets show that our
single-model face detector algorithm has similar or better performance compared
to the previous methods, which are more complex and require annotations of
either different poses or facial landmarks.
| no_new_dataset | 0.708452 |
1504.05150 | Mark Kaminski | Mark Kaminski, Bernardo Cuenca Grau | Computing Horn Rewritings of Description Logics Ontologies | 15 pages. To appear in IJCAI-15 | null | null | null | cs.AI cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the problem of rewriting an ontology O1 expressed in a DL L1 into an
ontology O2 in a Horn DL L2 such that O1 and O2 are equisatisfiable when
extended with an arbitrary dataset. Ontologies that admit such rewritings are
amenable to reasoning techniques ensuring tractability in data complexity.
After showing undecidability whenever L1 extends ALCF, we focus on devising
efficiently checkable conditions that ensure existence of a Horn rewriting. By
lifting existing techniques for rewriting Disjunctive Datalog programs into
plain Datalog to the case of arbitrary first-order programs with function
symbols, we identify a class of ontologies that admit Horn rewritings of
polynomial size. Our experiments indicate that many real-world ontologies
satisfy our sufficient conditions and thus admit polynomial Horn rewritings.
| [
{
"version": "v1",
"created": "Mon, 20 Apr 2015 18:39:27 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Apr 2015 10:59:25 GMT"
}
] | 2015-04-22T00:00:00 | [
[
"Kaminski",
"Mark",
""
],
[
"Grau",
"Bernardo Cuenca",
""
]
] | TITLE: Computing Horn Rewritings of Description Logics Ontologies
ABSTRACT: We study the problem of rewriting an ontology O1 expressed in a DL L1 into an
ontology O2 in a Horn DL L2 such that O1 and O2 are equisatisfiable when
extended with an arbitrary dataset. Ontologies that admit such rewritings are
amenable to reasoning techniques ensuring tractability in data complexity.
After showing undecidability whenever L1 extends ALCF, we focus on devising
efficiently checkable conditions that ensure existence of a Horn rewriting. By
lifting existing techniques for rewriting Disjunctive Datalog programs into
plain Datalog to the case of arbitrary first-order programs with function
symbols, we identify a class of ontologies that admit Horn rewritings of
polynomial size. Our experiments indicate that many real-world ontologies
satisfy our sufficient conditions and thus admit polynomial Horn rewritings.
| no_new_dataset | 0.709849 |
1504.05208 | Niclas Blomberg | Niclas Blomberg, Cristian R. Rojas and Bo Wahlberg | Approximate Regularization Paths for Nuclear Norm Minimization Using
Singular Value Bounds -- With Implementation and Extended Appendix | Also in conference version submitted to Signal Processing Education
Workshop 2015 | null | null | null | cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The widely used nuclear norm heuristic for rank minimization problems
introduces a regularization parameter which is difficult to tune. We have
recently proposed a method to approximate the regularization path, i.e., the
optimal solution as a function of the parameter, which requires solving the
problem only for a sparse set of points. In this paper, we extend the algorithm
to provide error bounds for the singular values of the approximation. We
exemplify the algorithms on large scale benchmark examples in model order
reduction. Here, the order of a dynamical system is reduced by means of
constrained minimization of the nuclear norm of a Hankel matrix.
| [
{
"version": "v1",
"created": "Mon, 20 Apr 2015 20:07:07 GMT"
}
] | 2015-04-22T00:00:00 | [
[
"Blomberg",
"Niclas",
""
],
[
"Rojas",
"Cristian R.",
""
],
[
"Wahlberg",
"Bo",
""
]
] | TITLE: Approximate Regularization Paths for Nuclear Norm Minimization Using
Singular Value Bounds -- With Implementation and Extended Appendix
ABSTRACT: The widely used nuclear norm heuristic for rank minimization problems
introduces a regularization parameter which is difficult to tune. We have
recently proposed a method to approximate the regularization path, i.e., the
optimal solution as a function of the parameter, which requires solving the
problem only for a sparse set of points. In this paper, we extend the algorithm
to provide error bounds for the singular values of the approximation. We
exemplify the algorithms on large scale benchmark examples in model order
reduction. Here, the order of a dynamical system is reduced by means of
constrained minimization of the nuclear norm of a Hankel matrix.
| no_new_dataset | 0.710377 |
1504.05473 | Yury Kashnitsky | Yury Kashnitsky, Dmitry I. Ignatov | Can FCA-based Recommender System Suggest a Proper Classifier? | 10 pages, 1 figure, 4 tables, ECAI 2014, workshop "What FCA can do
for "Artifficial Intelligence" | CEUR Workshop Proceedings, 1257, pp. 17-26 (2014) | null | null | cs.IR cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The paper briefly introduces multiple classifier systems and describes a new
algorithm, which improves classification accuracy by means of recommendation of
a proper algorithm to an object classification. This recommendation is done
assuming that a classifier is likely to predict the label of the object
correctly if it has correctly classified its neighbors. The process of
assigning a classifier to each object is based on Formal Concept Analysis. We
explain the idea of the algorithm with a toy example and describe our first
experiments with real-world datasets.
| [
{
"version": "v1",
"created": "Tue, 21 Apr 2015 15:38:23 GMT"
}
] | 2015-04-22T00:00:00 | [
[
"Kashnitsky",
"Yury",
""
],
[
"Ignatov",
"Dmitry I.",
""
]
] | TITLE: Can FCA-based Recommender System Suggest a Proper Classifier?
ABSTRACT: The paper briefly introduces multiple classifier systems and describes a new
algorithm, which improves classification accuracy by means of recommendation of
a proper algorithm to an object classification. This recommendation is done
assuming that a classifier is likely to predict the label of the object
correctly if it has correctly classified its neighbors. The process of
assigning a classifier to each object is based on Formal Concept Analysis. We
explain the idea of the algorithm with a toy example and describe our first
experiments with real-world datasets.
| no_new_dataset | 0.711049 |
1504.05524 | Dan Oneata | Heng Wang, Dan Oneata, Jakob Verbeek, Cordelia Schmid | A robust and efficient video representation for action recognition | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces a state-of-the-art video representation and applies it
to efficient action recognition and detection. We first propose to improve the
popular dense trajectory features by explicit camera motion estimation. More
specifically, we extract feature point matches between frames using SURF
descriptors and dense optical flow. The matches are used to estimate a
homography with RANSAC. To improve the robustness of homography estimation, a
human detector is employed to remove outlier matches from the human body as
human motion is not constrained by the camera. Trajectories consistent with the
homography are considered as due to camera motion, and thus removed. We also
use the homography to cancel out camera motion from the optical flow. This
results in significant improvement on motion-based HOF and MBH descriptors. We
further explore the recent Fisher vector as an alternative feature encoding
approach to the standard bag-of-words histogram, and consider different ways to
include spatial layout information in these encodings. We present a large and
varied set of evaluations, considering (i) classification of short basic
actions on six datasets, (ii) localization of such actions in feature-length
movies, and (iii) large-scale recognition of complex events. We find that our
improved trajectory features significantly outperform previous dense
trajectories, and that Fisher vectors are superior to bag-of-words encodings
for video recognition tasks. In all three tasks, we show substantial
improvements over the state-of-the-art results.
| [
{
"version": "v1",
"created": "Tue, 21 Apr 2015 17:44:07 GMT"
}
] | 2015-04-22T00:00:00 | [
[
"Wang",
"Heng",
""
],
[
"Oneata",
"Dan",
""
],
[
"Verbeek",
"Jakob",
""
],
[
"Schmid",
"Cordelia",
""
]
] | TITLE: A robust and efficient video representation for action recognition
ABSTRACT: This paper introduces a state-of-the-art video representation and applies it
to efficient action recognition and detection. We first propose to improve the
popular dense trajectory features by explicit camera motion estimation. More
specifically, we extract feature point matches between frames using SURF
descriptors and dense optical flow. The matches are used to estimate a
homography with RANSAC. To improve the robustness of homography estimation, a
human detector is employed to remove outlier matches from the human body as
human motion is not constrained by the camera. Trajectories consistent with the
homography are considered as due to camera motion, and thus removed. We also
use the homography to cancel out camera motion from the optical flow. This
results in significant improvement on motion-based HOF and MBH descriptors. We
further explore the recent Fisher vector as an alternative feature encoding
approach to the standard bag-of-words histogram, and consider different ways to
include spatial layout information in these encodings. We present a large and
varied set of evaluations, considering (i) classification of short basic
actions on six datasets, (ii) localization of such actions in feature-length
movies, and (iii) large-scale recognition of complex events. We find that our
improved trajectory features significantly outperform previous dense
trajectories, and that Fisher vectors are superior to bag-of-words encodings
for video recognition tasks. In all three tasks, we show substantial
improvements over the state-of-the-art results.
| no_new_dataset | 0.710465 |
1406.3407 | Gang Chen | Gang Chen and Sargur H. Srihari | Restricted Boltzmann Machine for Classification with Hierarchical
Correlated Prior | 13 pages, 5 figures | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-sa/3.0/ | Restricted Boltzmann machines (RBM) and its variants have become hot research
topics recently, and widely applied to many classification problems, such as
character recognition and document categorization. Often, classification RBM
ignores the interclass relationship or prior knowledge of sharing information
among classes. In this paper, we are interested in RBM with the hierarchical
prior over classes. We assume parameters for nearby nodes are correlated in the
hierarchical tree, and further the parameters at each node of the tree be
orthogonal to those at its ancestors. We propose a hierarchical correlated RBM
for classification problem, which generalizes the classification RBM with
sharing information among different classes. In order to reduce the redundancy
between node parameters in the hierarchy, we also introduce orthogonal
restrictions to our objective function. We test our method on challenge
datasets, and show promising results compared to competitive baselines.
| [
{
"version": "v1",
"created": "Fri, 13 Jun 2014 02:19:26 GMT"
},
{
"version": "v2",
"created": "Mon, 20 Apr 2015 18:39:18 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Chen",
"Gang",
""
],
[
"Srihari",
"Sargur H.",
""
]
] | TITLE: Restricted Boltzmann Machine for Classification with Hierarchical
Correlated Prior
ABSTRACT: Restricted Boltzmann machines (RBM) and its variants have become hot research
topics recently, and widely applied to many classification problems, such as
character recognition and document categorization. Often, classification RBM
ignores the interclass relationship or prior knowledge of sharing information
among classes. In this paper, we are interested in RBM with the hierarchical
prior over classes. We assume parameters for nearby nodes are correlated in the
hierarchical tree, and further the parameters at each node of the tree be
orthogonal to those at its ancestors. We propose a hierarchical correlated RBM
for classification problem, which generalizes the classification RBM with
sharing information among different classes. In order to reduce the redundancy
between node parameters in the hierarchy, we also introduce orthogonal
restrictions to our objective function. We test our method on challenge
datasets, and show promising results compared to competitive baselines.
| no_new_dataset | 0.712001 |
1406.5266 | Yaniv Taigman | Yaniv Taigman, Ming Yang, Marc'Aurelio Ranzato, Lior Wolf | Web-Scale Training for Face Identification | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/3.0/ | Scaling machine learning methods to very large datasets has attracted
considerable attention in recent years, thanks to easy access to ubiquitous
sensing and data from the web. We study face recognition and show that three
distinct properties have surprising effects on the transferability of deep
convolutional networks (CNN): (1) The bottleneck of the network serves as an
important transfer learning regularizer, and (2) in contrast to the common
wisdom, performance saturation may exist in CNN's (as the number of training
samples grows); we propose a solution for alleviating this by replacing the
naive random subsampling of the training set with a bootstrapping process.
Moreover, (3) we find a link between the representation norm and the ability to
discriminate in a target domain, which sheds lights on how such networks
represent faces. Based on these discoveries, we are able to improve face
recognition accuracy on the widely used LFW benchmark, both in the verification
(1:1) and identification (1:N) protocols, and directly compare, for the first
time, with the state of the art Commercially-Off-The-Shelf system and show a
sizable leap in performance.
| [
{
"version": "v1",
"created": "Fri, 20 Jun 2014 02:51:31 GMT"
},
{
"version": "v2",
"created": "Sat, 18 Apr 2015 09:18:19 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Taigman",
"Yaniv",
""
],
[
"Yang",
"Ming",
""
],
[
"Ranzato",
"Marc'Aurelio",
""
],
[
"Wolf",
"Lior",
""
]
] | TITLE: Web-Scale Training for Face Identification
ABSTRACT: Scaling machine learning methods to very large datasets has attracted
considerable attention in recent years, thanks to easy access to ubiquitous
sensing and data from the web. We study face recognition and show that three
distinct properties have surprising effects on the transferability of deep
convolutional networks (CNN): (1) The bottleneck of the network serves as an
important transfer learning regularizer, and (2) in contrast to the common
wisdom, performance saturation may exist in CNN's (as the number of training
samples grows); we propose a solution for alleviating this by replacing the
naive random subsampling of the training set with a bootstrapping process.
Moreover, (3) we find a link between the representation norm and the ability to
discriminate in a target domain, which sheds lights on how such networks
represent faces. Based on these discoveries, we are able to improve face
recognition accuracy on the widely used LFW benchmark, both in the verification
(1:1) and identification (1:N) protocols, and directly compare, for the first
time, with the state of the art Commercially-Off-The-Shelf system and show a
sizable leap in performance.
| no_new_dataset | 0.710245 |
1410.4355 | Erik Ferragut | Robert A. Bridges, John Collins, Erik M. Ferragut, Jason Laska, Blair
D. Sullivan | Multi-Level Anomaly Detection on Time-Varying Graph Data | 8 pages. Updated paper to address reviewer comments | null | null | null | cs.SI cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work presents a novel modeling and analysis framework for graph
sequences which addresses the challenge of detecting and contextualizing
anomalies in labelled, streaming graph data. We introduce a generalization of
the BTER model of Seshadhri et al. by adding flexibility to community
structure, and use this model to perform multi-scale graph anomaly detection.
Specifically, probability models describing coarse subgraphs are built by
aggregating probabilities at finer levels, and these closely related
hierarchical models simultaneously detect deviations from expectation. This
technique provides insight into a graph's structure and internal context that
may shed light on a detected event. Additionally, this multi-scale analysis
facilitates intuitive visualizations by allowing users to narrow focus from an
anomalous graph to particular subgraphs or nodes causing the anomaly.
For evaluation, two hierarchical anomaly detectors are tested against a
baseline Gaussian method on a series of sampled graphs. We demonstrate that our
graph statistics-based approach outperforms both a distribution-based detector
and the baseline in a labeled setting with community structure, and it
accurately detects anomalies in synthetic and real-world datasets at the node,
subgraph, and graph levels. To illustrate the accessibility of information made
possible via this technique, the anomaly detector and an associated interactive
visualization tool are tested on NCAA football data, where teams and
conferences that moved within the league are identified with perfect recall,
and precision greater than 0.786.
| [
{
"version": "v1",
"created": "Thu, 16 Oct 2014 09:57:20 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Oct 2014 19:08:37 GMT"
},
{
"version": "v3",
"created": "Fri, 17 Apr 2015 16:58:08 GMT"
},
{
"version": "v4",
"created": "Mon, 20 Apr 2015 11:55:53 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Bridges",
"Robert A.",
""
],
[
"Collins",
"John",
""
],
[
"Ferragut",
"Erik M.",
""
],
[
"Laska",
"Jason",
""
],
[
"Sullivan",
"Blair D.",
""
]
] | TITLE: Multi-Level Anomaly Detection on Time-Varying Graph Data
ABSTRACT: This work presents a novel modeling and analysis framework for graph
sequences which addresses the challenge of detecting and contextualizing
anomalies in labelled, streaming graph data. We introduce a generalization of
the BTER model of Seshadhri et al. by adding flexibility to community
structure, and use this model to perform multi-scale graph anomaly detection.
Specifically, probability models describing coarse subgraphs are built by
aggregating probabilities at finer levels, and these closely related
hierarchical models simultaneously detect deviations from expectation. This
technique provides insight into a graph's structure and internal context that
may shed light on a detected event. Additionally, this multi-scale analysis
facilitates intuitive visualizations by allowing users to narrow focus from an
anomalous graph to particular subgraphs or nodes causing the anomaly.
For evaluation, two hierarchical anomaly detectors are tested against a
baseline Gaussian method on a series of sampled graphs. We demonstrate that our
graph statistics-based approach outperforms both a distribution-based detector
and the baseline in a labeled setting with community structure, and it
accurately detects anomalies in synthetic and real-world datasets at the node,
subgraph, and graph levels. To illustrate the accessibility of information made
possible via this technique, the anomaly detector and an associated interactive
visualization tool are tested on NCAA football data, where teams and
conferences that moved within the league are identified with perfect recall,
and precision greater than 0.786.
| no_new_dataset | 0.711193 |
1412.6645 | Gabriel Synnaeve | Gabriel Synnaeve, Emmanuel Dupoux | Weakly Supervised Multi-Embeddings Learning of Acoustic Models | 6 pages, 3 figures | null | null | null | cs.SD cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We trained a Siamese network with multi-task same/different information on a
speech dataset, and found that it was possible to share a network for both
tasks without a loss in performance. The first task was to discriminate between
two same or different words, and the second was to discriminate between two
same or different talkers.
| [
{
"version": "v1",
"created": "Sat, 20 Dec 2014 11:54:41 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Feb 2015 10:09:09 GMT"
},
{
"version": "v3",
"created": "Mon, 20 Apr 2015 12:35:32 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Synnaeve",
"Gabriel",
""
],
[
"Dupoux",
"Emmanuel",
""
]
] | TITLE: Weakly Supervised Multi-Embeddings Learning of Acoustic Models
ABSTRACT: We trained a Siamese network with multi-task same/different information on a
speech dataset, and found that it was possible to share a network for both
tasks without a loss in performance. The first task was to discriminate between
two same or different words, and the second was to discriminate between two
same or different talkers.
| no_new_dataset | 0.708975 |
1501.06272 | Fang Zhao | Fang Zhao, Yongzhen Huang, Liang Wang, Tieniu Tan | Deep Semantic Ranking Based Hashing for Multi-Label Image Retrieval | CVPR 2015 | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the rapid growth of web images, hashing has received increasing
interests in large scale image retrieval. Research efforts have been devoted to
learning compact binary codes that preserve semantic similarity based on
labels. However, most of these hashing methods are designed to handle simple
binary similarity. The complex multilevel semantic structure of images
associated with multiple labels have not yet been well explored. Here we
propose a deep semantic ranking based method for learning hash functions that
preserve multilevel semantic similarity between multi-label images. In our
approach, deep convolutional neural network is incorporated into hash functions
to jointly learn feature representations and mappings from them to hash codes,
which avoids the limitation of semantic representation power of hand-crafted
features. Meanwhile, a ranking list that encodes the multilevel similarity
information is employed to guide the learning of such deep hash functions. An
effective scheme based on surrogate loss is used to solve the intractable
optimization problem of nonsmooth and multivariate ranking measures involved in
the learning procedure. Experimental results show the superiority of our
proposed approach over several state-of-the-art hashing methods in term of
ranking evaluation metrics when tested on multi-label image datasets.
| [
{
"version": "v1",
"created": "Mon, 26 Jan 2015 07:33:40 GMT"
},
{
"version": "v2",
"created": "Sun, 19 Apr 2015 04:28:58 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Zhao",
"Fang",
""
],
[
"Huang",
"Yongzhen",
""
],
[
"Wang",
"Liang",
""
],
[
"Tan",
"Tieniu",
""
]
] | TITLE: Deep Semantic Ranking Based Hashing for Multi-Label Image Retrieval
ABSTRACT: With the rapid growth of web images, hashing has received increasing
interests in large scale image retrieval. Research efforts have been devoted to
learning compact binary codes that preserve semantic similarity based on
labels. However, most of these hashing methods are designed to handle simple
binary similarity. The complex multilevel semantic structure of images
associated with multiple labels have not yet been well explored. Here we
propose a deep semantic ranking based method for learning hash functions that
preserve multilevel semantic similarity between multi-label images. In our
approach, deep convolutional neural network is incorporated into hash functions
to jointly learn feature representations and mappings from them to hash codes,
which avoids the limitation of semantic representation power of hand-crafted
features. Meanwhile, a ranking list that encodes the multilevel similarity
information is employed to guide the learning of such deep hash functions. An
effective scheme based on surrogate loss is used to solve the intractable
optimization problem of nonsmooth and multivariate ranking measures involved in
the learning procedure. Experimental results show the superiority of our
proposed approach over several state-of-the-art hashing methods in term of
ranking evaluation metrics when tested on multi-label image datasets.
| no_new_dataset | 0.709818 |
1504.04558 | Quanzeng You | Quanzeng You, Sumit Bhatia, Jiebo Luo | A Picture Tells a Thousand Words -- About You! User Interest Profiling
from User Generated Visual Content | 7 pages, 6 Figures, 4 Tables | null | null | null | cs.SI cs.IR cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Inference of online social network users' attributes and interests has been
an active research topic. Accurate identification of users' attributes and
interests is crucial for improving the performance of personalization and
recommender systems. Most of the existing works have focused on textual content
generated by the users and have successfully used it for predicting users'
interests and other identifying attributes. However, little attention has been
paid to user generated visual content (images) that is becoming increasingly
popular and pervasive in recent times. We posit that images posted by users on
online social networks are a reflection of topics they are interested in and
propose an approach to infer user attributes from images posted by them. We
analyze the content of individual images and then aggregate the image-level
knowledge to infer user-level interest distribution. We employ image-level
similarity to propagate the label information between images, as well as
utilize the image category information derived from the user created
organization structure to further propagate the category-level knowledge for
all images. A real life social network dataset created from Pinterest is used
for evaluation and the experimental results demonstrate the effectiveness of
our proposed approach.
| [
{
"version": "v1",
"created": "Fri, 17 Apr 2015 16:28:35 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"You",
"Quanzeng",
""
],
[
"Bhatia",
"Sumit",
""
],
[
"Luo",
"Jiebo",
""
]
] | TITLE: A Picture Tells a Thousand Words -- About You! User Interest Profiling
from User Generated Visual Content
ABSTRACT: Inference of online social network users' attributes and interests has been
an active research topic. Accurate identification of users' attributes and
interests is crucial for improving the performance of personalization and
recommender systems. Most of the existing works have focused on textual content
generated by the users and have successfully used it for predicting users'
interests and other identifying attributes. However, little attention has been
paid to user generated visual content (images) that is becoming increasingly
popular and pervasive in recent times. We posit that images posted by users on
online social networks are a reflection of topics they are interested in and
propose an approach to infer user attributes from images posted by them. We
analyze the content of individual images and then aggregate the image-level
knowledge to infer user-level interest distribution. We employ image-level
similarity to propagate the label information between images, as well as
utilize the image category information derived from the user created
organization structure to further propagate the category-level knowledge for
all images. A real life social network dataset created from Pinterest is used
for evaluation and the experimental results demonstrate the effectiveness of
our proposed approach.
| no_new_dataset | 0.696346 |
1504.04596 | Yadong Zhu | Yadong Zhu, Yanyan Lan, Jiafeng Guo, Xueqi Cheng | Structural Learning of Diverse Ranking | Discriminant Function, Diversity Feature, Learning Framework | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Relevance and diversity are both crucial criteria for an effective search
system. In this paper, we propose a unified learning framework for
simultaneously optimizing both relevance and diversity. Specifically, the
problem is formalized as a structural learning framework optimizing
Diversity-Correlated Evaluation Measures (DCEM), such as ERR-IA, a-NDCG and
NRBP. Within this framework, the discriminant function is defined to be a
bi-criteria objective maximizing the sum of the relevance scores and
dissimilarities (or diversity) among the documents. Relevance and diversity
features are utilized to define the relevance scores and dissimilarities,
respectively. Compared with traditional methods, the advantages of our approach
lie in that: (1) Directly optimizing DCEM as the loss function is more
fundamental for the task; (2) Our framework does not rely on explicit diversity
information such as subtopics, thus is more adaptive to real application; (3)
The representation of diversity as the feature-based scoring function is more
flexible to incorporate rich diversity-based features into the learning
framework. Extensive experiments on the public TREC datasets show that our
approach significantly outperforms state-of-the-art diversification approaches,
which validate the above advantages.
| [
{
"version": "v1",
"created": "Fri, 17 Apr 2015 18:23:16 GMT"
},
{
"version": "v2",
"created": "Mon, 20 Apr 2015 09:59:00 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Zhu",
"Yadong",
""
],
[
"Lan",
"Yanyan",
""
],
[
"Guo",
"Jiafeng",
""
],
[
"Cheng",
"Xueqi",
""
]
] | TITLE: Structural Learning of Diverse Ranking
ABSTRACT: Relevance and diversity are both crucial criteria for an effective search
system. In this paper, we propose a unified learning framework for
simultaneously optimizing both relevance and diversity. Specifically, the
problem is formalized as a structural learning framework optimizing
Diversity-Correlated Evaluation Measures (DCEM), such as ERR-IA, a-NDCG and
NRBP. Within this framework, the discriminant function is defined to be a
bi-criteria objective maximizing the sum of the relevance scores and
dissimilarities (or diversity) among the documents. Relevance and diversity
features are utilized to define the relevance scores and dissimilarities,
respectively. Compared with traditional methods, the advantages of our approach
lie in that: (1) Directly optimizing DCEM as the loss function is more
fundamental for the task; (2) Our framework does not rely on explicit diversity
information such as subtopics, thus is more adaptive to real application; (3)
The representation of diversity as the feature-based scoring function is more
flexible to incorporate rich diversity-based features into the learning
framework. Extensive experiments on the public TREC datasets show that our
approach significantly outperforms state-of-the-art diversification approaches,
which validate the above advantages.
| no_new_dataset | 0.709629 |
1504.04646 | Kwetishe Danjuma | Kwetishe Joro Danjuma | Performance Evaluation of Machine Learning Algorithms in Post-operative
Life Expectancy in the Lung Cancer Patients | 11 pages, 3 figures, 2 tables, ISSN (Print): 1694-0814 | ISSN
(Online): 1694-0784 | IJCSI International Journal of Computer Science Issues, Volume 12,
Issue 2, March 2015 | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The nature of clinical data makes it difficult to quickly select, tune and
apply machine learning algorithms to clinical prognosis. As a result, a lot of
time is spent searching for the most appropriate machine learning algorithms
applicable in clinical prognosis that contains either binary-valued or
multi-valued attributes. The study set out to identify and evaluate the
performance of machine learning classification schemes applied in clinical
prognosis of post-operative life expectancy in the lung cancer patients.
Multilayer Perceptron, J48, and the Naive Bayes algorithms were used to train
and test models on Thoracic Surgery datasets obtained from the University of
California Irvine machine learning repository. Stratified 10-fold
cross-validation was used to evaluate baseline performance accuracy of the
classifiers. The comparative analysis shows that multilayer perceptron
performed best with classification accuracy of 82.3%, J48 came out second with
classification accuracy of 81.8%, and Naive Bayes came out the worst with
classification accuracy of 74.4%. The quality and outcome of the chosen machine
learning algorithms depends on the ingenuity of the clinical miner.
| [
{
"version": "v1",
"created": "Fri, 17 Apr 2015 22:05:34 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Danjuma",
"Kwetishe Joro",
""
]
] | TITLE: Performance Evaluation of Machine Learning Algorithms in Post-operative
Life Expectancy in the Lung Cancer Patients
ABSTRACT: The nature of clinical data makes it difficult to quickly select, tune and
apply machine learning algorithms to clinical prognosis. As a result, a lot of
time is spent searching for the most appropriate machine learning algorithms
applicable in clinical prognosis that contains either binary-valued or
multi-valued attributes. The study set out to identify and evaluate the
performance of machine learning classification schemes applied in clinical
prognosis of post-operative life expectancy in the lung cancer patients.
Multilayer Perceptron, J48, and the Naive Bayes algorithms were used to train
and test models on Thoracic Surgery datasets obtained from the University of
California Irvine machine learning repository. Stratified 10-fold
cross-validation was used to evaluate baseline performance accuracy of the
classifiers. The comparative analysis shows that multilayer perceptron
performed best with classification accuracy of 82.3%, J48 came out second with
classification accuracy of 81.8%, and Naive Bayes came out the worst with
classification accuracy of 74.4%. The quality and outcome of the chosen machine
learning algorithms depends on the ingenuity of the clinical miner.
| no_new_dataset | 0.709655 |
1504.04730 | Sourav Bhattacharya | Sourav Bhattacharya and Otto Huhta and N. Asokan | LookAhead: Augmenting Crowdsourced Website Reputation Systems With
Predictive Modeling | 12 pages | null | null | null | cs.CR cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Unsafe websites consist of malicious as well as inappropriate sites, such as
those hosting questionable or offensive content. Website reputation systems are
intended to help ordinary users steer away from these unsafe sites. However,
the process of assigning safety ratings for websites typically involves humans.
Consequently it is time consuming, costly and not scalable. This has resulted
in two major problems: (i) a significant proportion of the web space remains
unrated and (ii) there is an unacceptable time lag before new websites are
rated.
In this paper, we show that by leveraging structural and content-based
properties of websites, it is possible to reliably and efficiently predict
their safety ratings, thereby mitigating both problems. We demonstrate the
effectiveness of our approach using four datasets of up to 90,000 websites. We
use ratings from Web of Trust (WOT), a popular crowdsourced web reputation
system, as ground truth. We propose a novel ensemble classification technique
that makes opportunistic use of available structural and content properties of
webpages to predict their eventual ratings in two dimensions used by WOT:
trustworthiness and child safety. Ours is the first classification system to
predict such subjective ratings and the same approach works equally well in
identifying malicious websites. Across all datasets, our classification
performs well with average F$_1$-score in the 74--90\% range.
| [
{
"version": "v1",
"created": "Sat, 18 Apr 2015 15:13:13 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Bhattacharya",
"Sourav",
""
],
[
"Huhta",
"Otto",
""
],
[
"Asokan",
"N.",
""
]
] | TITLE: LookAhead: Augmenting Crowdsourced Website Reputation Systems With
Predictive Modeling
ABSTRACT: Unsafe websites consist of malicious as well as inappropriate sites, such as
those hosting questionable or offensive content. Website reputation systems are
intended to help ordinary users steer away from these unsafe sites. However,
the process of assigning safety ratings for websites typically involves humans.
Consequently it is time consuming, costly and not scalable. This has resulted
in two major problems: (i) a significant proportion of the web space remains
unrated and (ii) there is an unacceptable time lag before new websites are
rated.
In this paper, we show that by leveraging structural and content-based
properties of websites, it is possible to reliably and efficiently predict
their safety ratings, thereby mitigating both problems. We demonstrate the
effectiveness of our approach using four datasets of up to 90,000 websites. We
use ratings from Web of Trust (WOT), a popular crowdsourced web reputation
system, as ground truth. We propose a novel ensemble classification technique
that makes opportunistic use of available structural and content properties of
webpages to predict their eventual ratings in two dimensions used by WOT:
trustworthiness and child safety. Ours is the first classification system to
predict such subjective ratings and the same approach works equally well in
identifying malicious websites. Across all datasets, our classification
performs well with average F$_1$-score in the 74--90\% range.
| no_new_dataset | 0.710829 |
1504.04739 | Wojciech Czarnecki | Rafal Jozefowicz, Wojciech Marian Czarnecki | Fast optimization of Multithreshold Entropy Linear Classifier | Presented at Theoretical Foundations of Machine Learning 2015
(http://tfml.gmum.net), final version published in Schedae Informaticae
Journal | null | 10.4467/20838476SI.14.005.3022 | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multithreshold Entropy Linear Classifier (MELC) is a density based model
which searches for a linear projection maximizing the Cauchy-Schwarz Divergence
of dataset kernel density estimation. Despite its good empirical results, one
of its drawbacks is the optimization speed. In this paper we analyze how one
can speed it up through solving an approximate problem. We analyze two methods,
both similar to the approximate solutions of the Kernel Density Estimation
querying and provide adaptive schemes for selecting a crucial parameters based
on user-specified acceptable error. Furthermore we show how one can exploit
well known conjugate gradients and L-BFGS optimizers despite the fact that the
original optimization problem should be solved on the sphere. All above methods
and modifications are tested on 10 real life datasets from UCI repository to
confirm their practical usability.
| [
{
"version": "v1",
"created": "Sat, 18 Apr 2015 16:19:22 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Jozefowicz",
"Rafal",
""
],
[
"Czarnecki",
"Wojciech Marian",
""
]
] | TITLE: Fast optimization of Multithreshold Entropy Linear Classifier
ABSTRACT: Multithreshold Entropy Linear Classifier (MELC) is a density based model
which searches for a linear projection maximizing the Cauchy-Schwarz Divergence
of dataset kernel density estimation. Despite its good empirical results, one
of its drawbacks is the optimization speed. In this paper we analyze how one
can speed it up through solving an approximate problem. We analyze two methods,
both similar to the approximate solutions of the Kernel Density Estimation
querying and provide adaptive schemes for selecting a crucial parameters based
on user-specified acceptable error. Furthermore we show how one can exploit
well known conjugate gradients and L-BFGS optimizers despite the fact that the
original optimization problem should be solved on the sphere. All above methods
and modifications are tested on 10 real life datasets from UCI repository to
confirm their practical usability.
| no_new_dataset | 0.709856 |
1504.04740 | Wojciech Czarnecki | Wojciech Marian Czarnecki | On the consistency of Multithreshold Entropy Linear Classifier | Presented at Theoretical Foundations of Machine Learning 2015
(http://tfml.gmum.net), final version published in Schedae Informaticae
Journal | null | 10.4467/20838476SI.15.012.3034 | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multithreshold Entropy Linear Classifier (MELC) is a recent classifier idea
which employs information theoretic concept in order to create a multithreshold
maximum margin model. In this paper we analyze its consistency over
multithreshold linear models and show that its objective function upper bounds
the amount of misclassified points in a similar manner like hinge loss does in
support vector machines. For further confirmation we also conduct some
numerical experiments on five datasets.
| [
{
"version": "v1",
"created": "Sat, 18 Apr 2015 16:29:26 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Czarnecki",
"Wojciech Marian",
""
]
] | TITLE: On the consistency of Multithreshold Entropy Linear Classifier
ABSTRACT: Multithreshold Entropy Linear Classifier (MELC) is a recent classifier idea
which employs information theoretic concept in order to create a multithreshold
maximum margin model. In this paper we analyze its consistency over
multithreshold linear models and show that its objective function upper bounds
the amount of misclassified points in a similar manner like hinge loss does in
support vector machines. For further confirmation we also conduct some
numerical experiments on five datasets.
| no_new_dataset | 0.713257 |
1504.04788 | Wenlin Chen | Wenlin Chen and James T. Wilson and Stephen Tyree and Kilian Q.
Weinberger and Yixin Chen | Compressing Neural Networks with the Hashing Trick | null | null | null | null | cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As deep nets are increasingly used in applications suited for mobile devices,
a fundamental dilemma becomes apparent: the trend in deep learning is to grow
models to absorb ever-increasing data set sizes; however mobile devices are
designed with very little memory and cannot store such large models. We present
a novel network architecture, HashedNets, that exploits inherent redundancy in
neural networks to achieve drastic reductions in model sizes. HashedNets uses a
low-cost hash function to randomly group connection weights into hash buckets,
and all connections within the same hash bucket share a single parameter value.
These parameters are tuned to adjust to the HashedNets weight sharing
architecture with standard backprop during training. Our hashing procedure
introduces no additional memory overhead, and we demonstrate on several
benchmark data sets that HashedNets shrink the storage requirements of neural
networks substantially while mostly preserving generalization performance.
| [
{
"version": "v1",
"created": "Sun, 19 Apr 2015 04:24:15 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Chen",
"Wenlin",
""
],
[
"Wilson",
"James T.",
""
],
[
"Tyree",
"Stephen",
""
],
[
"Weinberger",
"Kilian Q.",
""
],
[
"Chen",
"Yixin",
""
]
] | TITLE: Compressing Neural Networks with the Hashing Trick
ABSTRACT: As deep nets are increasingly used in applications suited for mobile devices,
a fundamental dilemma becomes apparent: the trend in deep learning is to grow
models to absorb ever-increasing data set sizes; however mobile devices are
designed with very little memory and cannot store such large models. We present
a novel network architecture, HashedNets, that exploits inherent redundancy in
neural networks to achieve drastic reductions in model sizes. HashedNets uses a
low-cost hash function to randomly group connection weights into hash buckets,
and all connections within the same hash bucket share a single parameter value.
These parameters are tuned to adjust to the HashedNets weight sharing
architecture with standard backprop during training. Our hashing procedure
introduces no additional memory overhead, and we demonstrate on several
benchmark data sets that HashedNets shrink the storage requirements of neural
networks substantially while mostly preserving generalization performance.
| no_new_dataset | 0.708799 |
1504.04871 | Sukrit Shankar | Sukrit Shankar, Vikas K. Garg, Roberto Cipolla | DEEP-CARVING: Discovering Visual Attributes by Carving Deep Neural Nets | 10 pages, 8 figures, CVPR 2015 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most of the approaches for discovering visual attributes in images demand
significant supervision, which is cumbersome to obtain. In this paper, we aim
to discover visual attributes in a weakly supervised setting that is commonly
encountered with contemporary image search engines. Deep Convolutional Neural
Networks (CNNs) have enjoyed remarkable success in vision applications
recently. However, in a weakly supervised scenario, widely used CNN training
procedures do not learn a robust model for predicting multiple attribute labels
simultaneously. The primary reason is that the attributes highly co-occur
within the training data. To ameliorate this limitation, we propose
Deep-Carving, a novel training procedure with CNNs, that helps the net
efficiently carve itself for the task of multiple attribute prediction. During
training, the responses of the feature maps are exploited in an ingenious way
to provide the net with multiple pseudo-labels (for training images) for
subsequent iterations. The process is repeated periodically after a fixed
number of iterations, and enables the net carve itself iteratively for
efficiently disentangling features. Additionally, we contribute a
noun-adjective pairing inspired Natural Scenes Attributes Dataset to the
research community, CAMIT - NSAD, containing a number of co-occurring
attributes within a noun category. We describe, in detail, salient aspects of
this dataset. Our experiments on CAMIT-NSAD and the SUN Attributes Dataset,
with weak supervision, clearly demonstrate that the Deep-Carved CNNs
consistently achieve considerable improvement in the precision of attribute
prediction over popular baseline methods.
| [
{
"version": "v1",
"created": "Sun, 19 Apr 2015 18:56:52 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Shankar",
"Sukrit",
""
],
[
"Garg",
"Vikas K.",
""
],
[
"Cipolla",
"Roberto",
""
]
] | TITLE: DEEP-CARVING: Discovering Visual Attributes by Carving Deep Neural Nets
ABSTRACT: Most of the approaches for discovering visual attributes in images demand
significant supervision, which is cumbersome to obtain. In this paper, we aim
to discover visual attributes in a weakly supervised setting that is commonly
encountered with contemporary image search engines. Deep Convolutional Neural
Networks (CNNs) have enjoyed remarkable success in vision applications
recently. However, in a weakly supervised scenario, widely used CNN training
procedures do not learn a robust model for predicting multiple attribute labels
simultaneously. The primary reason is that the attributes highly co-occur
within the training data. To ameliorate this limitation, we propose
Deep-Carving, a novel training procedure with CNNs, that helps the net
efficiently carve itself for the task of multiple attribute prediction. During
training, the responses of the feature maps are exploited in an ingenious way
to provide the net with multiple pseudo-labels (for training images) for
subsequent iterations. The process is repeated periodically after a fixed
number of iterations, and enables the net carve itself iteratively for
efficiently disentangling features. Additionally, we contribute a
noun-adjective pairing inspired Natural Scenes Attributes Dataset to the
research community, CAMIT - NSAD, containing a number of co-occurring
attributes within a noun category. We describe, in detail, salient aspects of
this dataset. Our experiments on CAMIT-NSAD and the SUN Attributes Dataset,
with weak supervision, clearly demonstrate that the Deep-Carved CNNs
consistently achieve considerable improvement in the precision of attribute
prediction over popular baseline methods.
| no_new_dataset | 0.607139 |
1504.04923 | Chunhua Shen | Ruizhi Qiao, Lingqiao Liu, Chunhua Shen, and Anton von den Hengel | Learning discriminative trajectorylet detector sets for accurate
skeleton-based action recognition | 10 pages | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The introduction of low-cost RGB-D sensors has promoted the research in
skeleton-based human action recognition. Devising a representation suitable for
characterising actions on the basis of noisy skeleton sequences remains a
challenge, however. We here provide two insights into this challenge. First, we
show that the discriminative information of a skeleton sequence usually resides
in a short temporal interval and we propose a simple-but-effective local
descriptor called trajectorylet to capture the static and kinematic information
within this interval. Second, we further propose to encode each trajectorylet
with a discriminative trajectorylet detector set which is selected from a large
number of candidate detectors trained through exemplar-SVMs. The action-level
representation is obtained by pooling trajectorylet encodings. Evaluating on
standard datasets acquired from the Kinect sensor, it is demonstrated that our
method obtains superior results over existing approaches under various
experimental setups.
| [
{
"version": "v1",
"created": "Mon, 20 Apr 2015 02:41:03 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Qiao",
"Ruizhi",
""
],
[
"Liu",
"Lingqiao",
""
],
[
"Shen",
"Chunhua",
""
],
[
"Hengel",
"Anton von den",
""
]
] | TITLE: Learning discriminative trajectorylet detector sets for accurate
skeleton-based action recognition
ABSTRACT: The introduction of low-cost RGB-D sensors has promoted the research in
skeleton-based human action recognition. Devising a representation suitable for
characterising actions on the basis of noisy skeleton sequences remains a
challenge, however. We here provide two insights into this challenge. First, we
show that the discriminative information of a skeleton sequence usually resides
in a short temporal interval and we propose a simple-but-effective local
descriptor called trajectorylet to capture the static and kinematic information
within this interval. Second, we further propose to encode each trajectorylet
with a discriminative trajectorylet detector set which is selected from a large
number of candidate detectors trained through exemplar-SVMs. The action-level
representation is obtained by pooling trajectorylet encodings. Evaluating on
standard datasets acquired from the Kinect sensor, it is demonstrated that our
method obtains superior results over existing approaches under various
experimental setups.
| no_new_dataset | 0.708591 |
1504.04945 | Yadong Zhu | Yadong Zhu, Yanyan Lan, Jiafeng Guo, Xueqi Cheng | Topic-focused Dynamic Information Filtering in Social Media | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the quick development of online social media such as twitter or sina
weibo in china, many users usually track hot topics to satisfy their desired
information need. For a hot topic, new opinions or ideas will be continuously
produced in the form of online data stream. In this scenario, how to
effectively filter and display information for a certain topic dynamically,
will be a critical problem. We call the problem as Topic-focused Dynamic
Information Filtering (denoted as TDIF for short) in social media. In this
paper, we start open discussions on such application problems. We first analyze
the properties of the TDIF problem, which usually contains several typical
requirements: relevance, diversity, recency and confidence. Recency means that
users want to follow the recent opinions or news. Additionally, the confidence
of information must be taken into consideration. How to balance these factors
properly in online data stream is very important and challenging. We propose a
dynamic preservation strategy on the basis of an existing feature-based utility
function, to solve the TDIF problem. Additionally, we propose new dynamic
diversity measures, to get a more reasonable evaluation for such application
problems. Extensive exploratory experiments have been conducted on TREC public
twitter dataset, and the experimental results validate the effectiveness of our
approach.
| [
{
"version": "v1",
"created": "Mon, 20 Apr 2015 06:12:06 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Zhu",
"Yadong",
""
],
[
"Lan",
"Yanyan",
""
],
[
"Guo",
"Jiafeng",
""
],
[
"Cheng",
"Xueqi",
""
]
] | TITLE: Topic-focused Dynamic Information Filtering in Social Media
ABSTRACT: With the quick development of online social media such as twitter or sina
weibo in china, many users usually track hot topics to satisfy their desired
information need. For a hot topic, new opinions or ideas will be continuously
produced in the form of online data stream. In this scenario, how to
effectively filter and display information for a certain topic dynamically,
will be a critical problem. We call the problem as Topic-focused Dynamic
Information Filtering (denoted as TDIF for short) in social media. In this
paper, we start open discussions on such application problems. We first analyze
the properties of the TDIF problem, which usually contains several typical
requirements: relevance, diversity, recency and confidence. Recency means that
users want to follow the recent opinions or news. Additionally, the confidence
of information must be taken into consideration. How to balance these factors
properly in online data stream is very important and challenging. We propose a
dynamic preservation strategy on the basis of an existing feature-based utility
function, to solve the TDIF problem. Additionally, we propose new dynamic
diversity measures, to get a more reasonable evaluation for such application
problems. Extensive exploratory experiments have been conducted on TREC public
twitter dataset, and the experimental results validate the effectiveness of our
approach.
| no_new_dataset | 0.711825 |
1504.05035 | Wangmeng Zuo | Xiaohe Wu, Wangmeng Zuo, Yuanyuan Zhu, Liang Lin | F-SVM: Combination of Feature Transformation and SVM Learning via Convex
Relaxation | 11 pages, 5 figures | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The generalization error bound of support vector machine (SVM) depends on the
ratio of radius and margin, while standard SVM only considers the maximization
of the margin but ignores the minimization of the radius. Several approaches
have been proposed to integrate radius and margin for joint learning of feature
transformation and SVM classifier. However, most of them either require the
form of the transformation matrix to be diagonal, or are non-convex and
computationally expensive. In this paper, we suggest a novel approximation for
the radius of minimum enclosing ball (MEB) in feature space, and then propose a
convex radius-margin based SVM model for joint learning of feature
transformation and SVM classifier, i.e., F-SVM. An alternating minimization
method is adopted to solve the F-SVM model, where the feature transformation is
updatedvia gradient descent and the classifier is updated by employing the
existing SVM solver. By incorporating with kernel principal component analysis,
F-SVM is further extended for joint learning of nonlinear transformation and
classifier. Experimental results on the UCI machine learning datasets and the
LFW face datasets show that F-SVM outperforms the standard SVM and the existing
radius-margin based SVMs, e.g., RMM, R-SVM+ and R-SVM+{\mu}.
| [
{
"version": "v1",
"created": "Mon, 20 Apr 2015 12:36:50 GMT"
}
] | 2015-04-21T00:00:00 | [
[
"Wu",
"Xiaohe",
""
],
[
"Zuo",
"Wangmeng",
""
],
[
"Zhu",
"Yuanyuan",
""
],
[
"Lin",
"Liang",
""
]
] | TITLE: F-SVM: Combination of Feature Transformation and SVM Learning via Convex
Relaxation
ABSTRACT: The generalization error bound of support vector machine (SVM) depends on the
ratio of radius and margin, while standard SVM only considers the maximization
of the margin but ignores the minimization of the radius. Several approaches
have been proposed to integrate radius and margin for joint learning of feature
transformation and SVM classifier. However, most of them either require the
form of the transformation matrix to be diagonal, or are non-convex and
computationally expensive. In this paper, we suggest a novel approximation for
the radius of minimum enclosing ball (MEB) in feature space, and then propose a
convex radius-margin based SVM model for joint learning of feature
transformation and SVM classifier, i.e., F-SVM. An alternating minimization
method is adopted to solve the F-SVM model, where the feature transformation is
updatedvia gradient descent and the classifier is updated by employing the
existing SVM solver. By incorporating with kernel principal component analysis,
F-SVM is further extended for joint learning of nonlinear transformation and
classifier. Experimental results on the UCI machine learning datasets and the
LFW face datasets show that F-SVM outperforms the standard SVM and the existing
radius-margin based SVMs, e.g., RMM, R-SVM+ and R-SVM+{\mu}.
| no_new_dataset | 0.712664 |
1305.1495 | Lorenzo Del Castello | Alessandro Attanasi, Andrea Cavagna, Lorenzo Del Castello, Irene
Giardina, Asja Jelic, Stefania Melillo, Leonardo Parisi, Fabio Pellacini,
Edward Shen, Edmondo Silvestri, Massimiliano Viale | GReTA - a novel Global and Recursive Tracking Algorithm in three
dimensions | 13 pages, 6 figures, 3 tables. Version 3 was slightly shortened, and
new comprative results on the public datasets (thermal infrared videos of
flying bats) by Z. Wu and coworkers (2014) were included. in A. Attanasi et
al., "GReTA - A Novel Global and Recursive Tracking Algorithm in Three
Dimensions", IEEE Trans. Pattern Anal. Mach. Intell., vol.37 (2015) | null | 10.1109/TPAMI.2015.2414427 | null | q-bio.QM cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Tracking multiple moving targets allows quantitative measure of the dynamic
behavior in systems as diverse as animal groups in biology, turbulence in fluid
dynamics and crowd and traffic control. In three dimensions, tracking several
targets becomes increasingly hard since optical occlusions are very likely,
i.e. two featureless targets frequently overlap for several frames. Occlusions
are particularly frequent in biological groups such as bird flocks, fish
schools, and insect swarms, a fact that has severely limited collective animal
behavior field studies in the past. This paper presents a 3D tracking method
that is robust in the case of severe occlusions. To ensure robustness, we adopt
a global optimization approach that works on all objects and frames at once. To
achieve practicality and scalability, we employ a divide and conquer
formulation, thanks to which the computational complexity of the problem is
reduced by orders of magnitude. We tested our algorithm with synthetic data,
with experimental data of bird flocks and insect swarms and with public
benchmark datasets, and show that our system yields high quality trajectories
for hundreds of moving targets with severe overlap. The results obtained on
very heterogeneous data show the potential applicability of our method to the
most diverse experimental situations.
| [
{
"version": "v1",
"created": "Tue, 7 May 2013 12:45:30 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Apr 2014 14:55:59 GMT"
},
{
"version": "v3",
"created": "Fri, 17 Apr 2015 16:36:59 GMT"
}
] | 2015-04-20T00:00:00 | [
[
"Attanasi",
"Alessandro",
""
],
[
"Cavagna",
"Andrea",
""
],
[
"Del Castello",
"Lorenzo",
""
],
[
"Giardina",
"Irene",
""
],
[
"Jelic",
"Asja",
""
],
[
"Melillo",
"Stefania",
""
],
[
"Parisi",
"Leonardo",
""
],
[
"Pellacini",
"Fabio",
""
],
[
"Shen",
"Edward",
""
],
[
"Silvestri",
"Edmondo",
""
],
[
"Viale",
"Massimiliano",
""
]
] | TITLE: GReTA - a novel Global and Recursive Tracking Algorithm in three
dimensions
ABSTRACT: Tracking multiple moving targets allows quantitative measure of the dynamic
behavior in systems as diverse as animal groups in biology, turbulence in fluid
dynamics and crowd and traffic control. In three dimensions, tracking several
targets becomes increasingly hard since optical occlusions are very likely,
i.e. two featureless targets frequently overlap for several frames. Occlusions
are particularly frequent in biological groups such as bird flocks, fish
schools, and insect swarms, a fact that has severely limited collective animal
behavior field studies in the past. This paper presents a 3D tracking method
that is robust in the case of severe occlusions. To ensure robustness, we adopt
a global optimization approach that works on all objects and frames at once. To
achieve practicality and scalability, we employ a divide and conquer
formulation, thanks to which the computational complexity of the problem is
reduced by orders of magnitude. We tested our algorithm with synthetic data,
with experimental data of bird flocks and insect swarms and with public
benchmark datasets, and show that our system yields high quality trajectories
for hundreds of moving targets with severe overlap. The results obtained on
very heterogeneous data show the potential applicability of our method to the
most diverse experimental situations.
| no_new_dataset | 0.711268 |
1408.5661 | Keisuke Yamazaki | Keisuke Yamazaki | Asymptotic Accuracy of Bayesian Estimation for a Single Latent Variable | 28 pages, 3 figures | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In data science and machine learning, hierarchical parametric models, such as
mixture models, are often used. They contain two kinds of variables: observable
variables, which represent the parts of the data that can be directly measured,
and latent variables, which represent the underlying processes that generate
the data. Although there has been an increase in research on the estimation
accuracy for observable variables, the theoretical analysis of estimating
latent variables has not been thoroughly investigated. In a previous study, we
determined the accuracy of a Bayes estimation for the joint probability of the
latent variables in a dataset, and we proved that the Bayes method is
asymptotically more accurate than the maximum-likelihood method. However, the
accuracy of the Bayes estimation for a single latent variable remains unknown.
In the present paper, we derive the asymptotic expansions of the error
functions, which are defined by the Kullback-Leibler divergence, for two types
of single-variable estimations when the statistical regularity is satisfied.
Our results indicate that the accuracies of the Bayes and maximum-likelihood
methods are asymptotically equivalent and clarify that the Bayes method is only
advantageous for multivariable estimations.
| [
{
"version": "v1",
"created": "Mon, 25 Aug 2014 04:44:53 GMT"
},
{
"version": "v2",
"created": "Tue, 27 Jan 2015 04:00:06 GMT"
},
{
"version": "v3",
"created": "Fri, 17 Apr 2015 06:59:26 GMT"
}
] | 2015-04-20T00:00:00 | [
[
"Yamazaki",
"Keisuke",
""
]
] | TITLE: Asymptotic Accuracy of Bayesian Estimation for a Single Latent Variable
ABSTRACT: In data science and machine learning, hierarchical parametric models, such as
mixture models, are often used. They contain two kinds of variables: observable
variables, which represent the parts of the data that can be directly measured,
and latent variables, which represent the underlying processes that generate
the data. Although there has been an increase in research on the estimation
accuracy for observable variables, the theoretical analysis of estimating
latent variables has not been thoroughly investigated. In a previous study, we
determined the accuracy of a Bayes estimation for the joint probability of the
latent variables in a dataset, and we proved that the Bayes method is
asymptotically more accurate than the maximum-likelihood method. However, the
accuracy of the Bayes estimation for a single latent variable remains unknown.
In the present paper, we derive the asymptotic expansions of the error
functions, which are defined by the Kullback-Leibler divergence, for two types
of single-variable estimations when the statistical regularity is satisfied.
Our results indicate that the accuracies of the Bayes and maximum-likelihood
methods are asymptotically equivalent and clarify that the Bayes method is only
advantageous for multivariable estimations.
| no_new_dataset | 0.710057 |
1410.8151 | Ahmet Iscen | Ahmet Iscen, Giorgos Tolias, Philippe-Henri Gosselin, Herv\'e J\'egou | A comparison of dense region detectors for image search and fine-grained
classification | Accepted to IEEE Transactions on Image Processing | null | 10.1109/TIP.2015.2423557 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider a pipeline for image classification or search based on coding
approaches like Bag of Words or Fisher vectors. In this context, the most
common approach is to extract the image patches regularly in a dense manner on
several scales. This paper proposes and evaluates alternative choices to
extract patches densely. Beyond simple strategies derived from regular interest
region detectors, we propose approaches based on super-pixels, edges, and a
bank of Zernike filters used as detectors. The different approaches are
evaluated on recent image retrieval and fine-grain classification benchmarks.
Our results show that the regular dense detector is outperformed by other
methods in most situations, leading us to improve the state of the art in
comparable setups on standard retrieval and fined-grain benchmarks. As a
byproduct of our study, we show that existing methods for blob and super-pixel
extraction achieve high accuracy if the patches are extracted along the edges
and not around the detected regions.
| [
{
"version": "v1",
"created": "Wed, 29 Oct 2014 20:18:41 GMT"
},
{
"version": "v2",
"created": "Thu, 26 Feb 2015 12:56:39 GMT"
},
{
"version": "v3",
"created": "Fri, 17 Apr 2015 18:36:17 GMT"
}
] | 2015-04-20T00:00:00 | [
[
"Iscen",
"Ahmet",
""
],
[
"Tolias",
"Giorgos",
""
],
[
"Gosselin",
"Philippe-Henri",
""
],
[
"Jégou",
"Hervé",
""
]
] | TITLE: A comparison of dense region detectors for image search and fine-grained
classification
ABSTRACT: We consider a pipeline for image classification or search based on coding
approaches like Bag of Words or Fisher vectors. In this context, the most
common approach is to extract the image patches regularly in a dense manner on
several scales. This paper proposes and evaluates alternative choices to
extract patches densely. Beyond simple strategies derived from regular interest
region detectors, we propose approaches based on super-pixels, edges, and a
bank of Zernike filters used as detectors. The different approaches are
evaluated on recent image retrieval and fine-grain classification benchmarks.
Our results show that the regular dense detector is outperformed by other
methods in most situations, leading us to improve the state of the art in
comparable setups on standard retrieval and fined-grain benchmarks. As a
byproduct of our study, we show that existing methods for blob and super-pixel
extraction achieve high accuracy if the patches are extracted along the edges
and not around the detected regions.
| no_new_dataset | 0.712207 |
1412.5083 | Qiang Qiu | Qiang Qiu, Guillermo Sapiro, Alex Bronstein | Random Forests Can Hash | null | null | null | null | cs.CV cs.IR cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hash codes are a very efficient data representation needed to be able to cope
with the ever growing amounts of data. We introduce a random forest semantic
hashing scheme with information-theoretic code aggregation, showing for the
first time how random forest, a technique that together with deep learning have
shown spectacular results in classification, can also be extended to
large-scale retrieval. Traditional random forest fails to enforce the
consistency of hashes generated from each tree for the same class data, i.e.,
to preserve the underlying similarity, and it also lacks a principled way for
code aggregation across trees. We start with a simple hashing scheme, where
independently trained random trees in a forest are acting as hashing functions.
We the propose a subspace model as the splitting function, and show that it
enforces the hash consistency in a tree for data from the same class. We also
introduce an information-theoretic approach for aggregating codes of individual
trees into a single hash code, producing a near-optimal unique hash for each
class. Experiments on large-scale public datasets are presented, showing that
the proposed approach significantly outperforms state-of-the-art hashing
methods for retrieval tasks.
| [
{
"version": "v1",
"created": "Tue, 16 Dec 2014 17:02:18 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Feb 2015 18:26:12 GMT"
},
{
"version": "v3",
"created": "Fri, 17 Apr 2015 01:00:24 GMT"
}
] | 2015-04-20T00:00:00 | [
[
"Qiu",
"Qiang",
""
],
[
"Sapiro",
"Guillermo",
""
],
[
"Bronstein",
"Alex",
""
]
] | TITLE: Random Forests Can Hash
ABSTRACT: Hash codes are a very efficient data representation needed to be able to cope
with the ever growing amounts of data. We introduce a random forest semantic
hashing scheme with information-theoretic code aggregation, showing for the
first time how random forest, a technique that together with deep learning have
shown spectacular results in classification, can also be extended to
large-scale retrieval. Traditional random forest fails to enforce the
consistency of hashes generated from each tree for the same class data, i.e.,
to preserve the underlying similarity, and it also lacks a principled way for
code aggregation across trees. We start with a simple hashing scheme, where
independently trained random trees in a forest are acting as hashing functions.
We the propose a subspace model as the splitting function, and show that it
enforces the hash consistency in a tree for data from the same class. We also
introduce an information-theoretic approach for aggregating codes of individual
trees into a single hash code, producing a near-optimal unique hash for each
class. Experiments on large-scale public datasets are presented, showing that
the proposed approach significantly outperforms state-of-the-art hashing
methods for retrieval tasks.
| no_new_dataset | 0.710672 |
1503.05787 | Tobias Isenberg | Maarten H. Everts, Henk Bekker, Jos B.T.M. Roerdink, and Tobias
Isenberg | Interactive Illustrative Line Styles and Line Style Transfer Functions
for Flow Visualization | Extended version of a short paper at Pacific Graphics 2011
(http://dx.doi.org/10.2312/PE/PG/PG2011short/105-110) | null | null | null | cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a flexible illustrative line style model for the visualization of
streamline data. Our model partitions view-oriented line strips into parallel
bands whose basic visual properties can be controlled independently. We thus
extend previous line stylization techniques specifically for visualization
purposes by allowing the parametrization of these bands based on the local line
data attributes. Moreover, our approach supports emphasis and abstraction by
introducing line style transfer functions that map local line attribute values
to complete line styles. With a flexible GPU implementation of this line style
model we enable the interactive exploration of visual representations of
streamlines. We demonstrate the effectiveness of our model by applying it to 3D
flow field datasets.
| [
{
"version": "v1",
"created": "Thu, 19 Mar 2015 14:53:55 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Apr 2015 14:01:50 GMT"
}
] | 2015-04-20T00:00:00 | [
[
"Everts",
"Maarten H.",
""
],
[
"Bekker",
"Henk",
""
],
[
"Roerdink",
"Jos B. T. M.",
""
],
[
"Isenberg",
"Tobias",
""
]
] | TITLE: Interactive Illustrative Line Styles and Line Style Transfer Functions
for Flow Visualization
ABSTRACT: We present a flexible illustrative line style model for the visualization of
streamline data. Our model partitions view-oriented line strips into parallel
bands whose basic visual properties can be controlled independently. We thus
extend previous line stylization techniques specifically for visualization
purposes by allowing the parametrization of these bands based on the local line
data attributes. Moreover, our approach supports emphasis and abstraction by
introducing line style transfer functions that map local line attribute values
to complete line styles. With a flexible GPU implementation of this line style
model we enable the interactive exploration of visual representations of
streamlines. We demonstrate the effectiveness of our model by applying it to 3D
flow field datasets.
| no_new_dataset | 0.710258 |
1504.04531 | Nicolas Dobigeon | Laetitia Loncan, Luis B. Almeida, Jos\'e M. Bioucas-Dias, Xavier
Briottet, Jocelyn Chanussot, Nicolas Dobigeon, Sophie Fabre, Wenzhi Liao,
Giorgio A. Licciardi, Miguel Sim\~oes, Jean-Yves Tourneret, Miguel A.
Veganzones, Gemine Vivone, Qi Wei and Naoto Yokoya | Hyperspectral pansharpening: a review | null | null | null | null | cs.CV physics.data-an stat.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Pansharpening aims at fusing a panchromatic image with a multispectral one,
to generate an image with the high spatial resolution of the former and the
high spectral resolution of the latter. In the last decade, many algorithms
have been presented in the literature for pansharpening using multispectral
data. With the increasing availability of hyperspectral systems, these methods
are now being adapted to hyperspectral images. In this work, we compare new
pansharpening techniques designed for hyperspectral data with some of the state
of the art methods for multispectral pansharpening, which have been adapted for
hyperspectral data. Eleven methods from different classes (component
substitution, multiresolution analysis, hybrid, Bayesian and matrix
factorization) are analyzed. These methods are applied to three datasets and
their effectiveness and robustness are evaluated with widely used performance
indicators. In addition, all the pansharpening techniques considered in this
paper have been implemented in a MATLAB toolbox that is made available to the
community.
| [
{
"version": "v1",
"created": "Fri, 17 Apr 2015 15:07:11 GMT"
}
] | 2015-04-20T00:00:00 | [
[
"Loncan",
"Laetitia",
""
],
[
"Almeida",
"Luis B.",
""
],
[
"Bioucas-Dias",
"José M.",
""
],
[
"Briottet",
"Xavier",
""
],
[
"Chanussot",
"Jocelyn",
""
],
[
"Dobigeon",
"Nicolas",
""
],
[
"Fabre",
"Sophie",
""
],
[
"Liao",
"Wenzhi",
""
],
[
"Licciardi",
"Giorgio A.",
""
],
[
"Simões",
"Miguel",
""
],
[
"Tourneret",
"Jean-Yves",
""
],
[
"Veganzones",
"Miguel A.",
""
],
[
"Vivone",
"Gemine",
""
],
[
"Wei",
"Qi",
""
],
[
"Yokoya",
"Naoto",
""
]
] | TITLE: Hyperspectral pansharpening: a review
ABSTRACT: Pansharpening aims at fusing a panchromatic image with a multispectral one,
to generate an image with the high spatial resolution of the former and the
high spectral resolution of the latter. In the last decade, many algorithms
have been presented in the literature for pansharpening using multispectral
data. With the increasing availability of hyperspectral systems, these methods
are now being adapted to hyperspectral images. In this work, we compare new
pansharpening techniques designed for hyperspectral data with some of the state
of the art methods for multispectral pansharpening, which have been adapted for
hyperspectral data. Eleven methods from different classes (component
substitution, multiresolution analysis, hybrid, Bayesian and matrix
factorization) are analyzed. These methods are applied to three datasets and
their effectiveness and robustness are evaluated with widely used performance
indicators. In addition, all the pansharpening techniques considered in this
paper have been implemented in a MATLAB toolbox that is made available to the
community.
| no_new_dataset | 0.711838 |
1504.04548 | Simone Bianco | Simone Bianco, Claudio Cusano, Raimondo Schettini | Color Constancy Using CNNs | Accepted at DeepVision: Deep Learning in Computer Vision 2015 (CVPR
2015 workshop) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work we describe a Convolutional Neural Network (CNN) to accurately
predict the scene illumination. Taking image patches as input, the CNN works in
the spatial domain without using hand-crafted features that are employed by
most previous methods. The network consists of one convolutional layer with max
pooling, one fully connected layer and three output nodes. Within the network
structure, feature learning and regression are integrated into one optimization
process, which leads to a more effective model for estimating scene
illumination. This approach achieves state-of-the-art performance on a standard
dataset of RAW images. Preliminary experiments on images with spatially varying
illumination demonstrate the stability of the local illuminant estimation
ability of our CNN.
| [
{
"version": "v1",
"created": "Fri, 17 Apr 2015 15:51:07 GMT"
}
] | 2015-04-20T00:00:00 | [
[
"Bianco",
"Simone",
""
],
[
"Cusano",
"Claudio",
""
],
[
"Schettini",
"Raimondo",
""
]
] | TITLE: Color Constancy Using CNNs
ABSTRACT: In this work we describe a Convolutional Neural Network (CNN) to accurately
predict the scene illumination. Taking image patches as input, the CNN works in
the spatial domain without using hand-crafted features that are employed by
most previous methods. The network consists of one convolutional layer with max
pooling, one fully connected layer and three output nodes. Within the network
structure, feature learning and regression are integrated into one optimization
process, which leads to a more effective model for estimating scene
illumination. This approach achieves state-of-the-art performance on a standard
dataset of RAW images. Preliminary experiments on images with spatially varying
illumination demonstrate the stability of the local illuminant estimation
ability of our CNN.
| no_new_dataset | 0.710377 |
1504.04588 | James-A. Goulet | James-A. Goulet | The Nataf-Beta Random Field Classifier: An Extension of the Beta
Conjugate Prior to Classification Problems | 17 pages, 4 figures, Submitted for publication in the Journal of
Machine Learning Research | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents the Nataf-Beta Random Field Classifier, a discriminative
approach that extends the applicability of the Beta conjugate prior to
classification problems. The approach's key feature is to model the probability
of a class conditional on attribute values as a random field whose marginals
are Beta distributed, and where the parameters of marginals are themselves
described by random fields. Although the classification accuracy of the
approach proposed does not statistically outperform the best accuracies
reported in the literature, it ranks among the top tier for the six benchmark
datasets tested. The Nataf-Beta Random Field Classifier is suited as a general
purpose classification approach for real-continuous and real-integer attribute
value problems.
| [
{
"version": "v1",
"created": "Fri, 17 Apr 2015 17:32:00 GMT"
}
] | 2015-04-20T00:00:00 | [
[
"Goulet",
"James-A.",
""
]
] | TITLE: The Nataf-Beta Random Field Classifier: An Extension of the Beta
Conjugate Prior to Classification Problems
ABSTRACT: This paper presents the Nataf-Beta Random Field Classifier, a discriminative
approach that extends the applicability of the Beta conjugate prior to
classification problems. The approach's key feature is to model the probability
of a class conditional on attribute values as a random field whose marginals
are Beta distributed, and where the parameters of marginals are themselves
described by random fields. Although the classification accuracy of the
approach proposed does not statistically outperform the best accuracies
reported in the literature, it ranks among the top tier for the six benchmark
datasets tested. The Nataf-Beta Random Field Classifier is suited as a general
purpose classification approach for real-continuous and real-integer attribute
value problems.
| no_new_dataset | 0.711982 |
1409.0940 | Haim Avron | Vikas Sindhwani and Haim Avron | High-performance Kernel Machines with Implicit Distributed Optimization
and Randomization | Work presented at MMDS 2014 (June 2014) and JSM 2014 | null | null | null | stat.ML cs.DC cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In order to fully utilize "big data", it is often required to use "big
models". Such models tend to grow with the complexity and size of the training
data, and do not make strong parametric assumptions upfront on the nature of
the underlying statistical dependencies. Kernel methods fit this need well, as
they constitute a versatile and principled statistical methodology for solving
a wide range of non-parametric modelling problems. However, their high
computational costs (in storage and time) pose a significant barrier to their
widespread adoption in big data applications.
We propose an algorithmic framework and high-performance implementation for
massive-scale training of kernel-based statistical models, based on combining
two key technical ingredients: (i) distributed general purpose convex
optimization, and (ii) the use of randomization to improve the scalability of
kernel methods. Our approach is based on a block-splitting variant of the
Alternating Directions Method of Multipliers, carefully reconfigured to handle
very large random feature matrices, while exploiting hybrid parallelism
typically found in modern clusters of multicore machines. Our implementation
supports a variety of statistical learning tasks by enabling several loss
functions, regularization schemes, kernels, and layers of randomized
approximations for both dense and sparse datasets, in a highly extensible
framework. We evaluate the ability of our framework to learn models on data
from applications, and provide a comparison against existing sequential and
parallel libraries.
| [
{
"version": "v1",
"created": "Wed, 3 Sep 2014 02:28:51 GMT"
},
{
"version": "v2",
"created": "Tue, 23 Dec 2014 21:38:15 GMT"
},
{
"version": "v3",
"created": "Thu, 16 Apr 2015 18:06:53 GMT"
}
] | 2015-04-17T00:00:00 | [
[
"Sindhwani",
"Vikas",
""
],
[
"Avron",
"Haim",
""
]
] | TITLE: High-performance Kernel Machines with Implicit Distributed Optimization
and Randomization
ABSTRACT: In order to fully utilize "big data", it is often required to use "big
models". Such models tend to grow with the complexity and size of the training
data, and do not make strong parametric assumptions upfront on the nature of
the underlying statistical dependencies. Kernel methods fit this need well, as
they constitute a versatile and principled statistical methodology for solving
a wide range of non-parametric modelling problems. However, their high
computational costs (in storage and time) pose a significant barrier to their
widespread adoption in big data applications.
We propose an algorithmic framework and high-performance implementation for
massive-scale training of kernel-based statistical models, based on combining
two key technical ingredients: (i) distributed general purpose convex
optimization, and (ii) the use of randomization to improve the scalability of
kernel methods. Our approach is based on a block-splitting variant of the
Alternating Directions Method of Multipliers, carefully reconfigured to handle
very large random feature matrices, while exploiting hybrid parallelism
typically found in modern clusters of multicore machines. Our implementation
supports a variety of statistical learning tasks by enabling several loss
functions, regularization schemes, kernels, and layers of randomized
approximations for both dense and sparse datasets, in a highly extensible
framework. We evaluate the ability of our framework to learn models on data
from applications, and provide a comparison against existing sequential and
parallel libraries.
| no_new_dataset | 0.706981 |
1502.05337 | Emiliano De Cristofaro | Julien Freudiger and Emiliano De Cristofaro and Alex Brito | Controlled Data Sharing for Collaborative Predictive Blacklisting | A preliminary version of this paper appears in DIMVA 2015. This is
the full version. arXiv admin note: substantial text overlap with
arXiv:1403.2123 | null | null | null | cs.CR cs.NI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although sharing data across organizations is often advocated as a promising
way to enhance cybersecurity, collaborative initiatives are rarely put into
practice owing to confidentiality, trust, and liability challenges. In this
paper, we investigate whether collaborative threat mitigation can be realized
via a controlled data sharing approach, whereby organizations make informed
decisions as to whether or not, and how much, to share. Using appropriate
cryptographic tools, entities can estimate the benefits of collaboration and
agree on what to share in a privacy-preserving way, without having to disclose
their datasets. We focus on collaborative predictive blacklisting, i.e.,
forecasting attack sources based on one's logs and those contributed by other
organizations. We study the impact of different sharing strategies by
experimenting on a real-world dataset of two billion suspicious IP addresses
collected from Dshield over two months. We find that controlled data sharing
yields up to 105% accuracy improvement on average, while also reducing the
false positive rate.
| [
{
"version": "v1",
"created": "Wed, 18 Feb 2015 18:48:56 GMT"
},
{
"version": "v2",
"created": "Thu, 16 Apr 2015 09:09:16 GMT"
}
] | 2015-04-17T00:00:00 | [
[
"Freudiger",
"Julien",
""
],
[
"De Cristofaro",
"Emiliano",
""
],
[
"Brito",
"Alex",
""
]
] | TITLE: Controlled Data Sharing for Collaborative Predictive Blacklisting
ABSTRACT: Although sharing data across organizations is often advocated as a promising
way to enhance cybersecurity, collaborative initiatives are rarely put into
practice owing to confidentiality, trust, and liability challenges. In this
paper, we investigate whether collaborative threat mitigation can be realized
via a controlled data sharing approach, whereby organizations make informed
decisions as to whether or not, and how much, to share. Using appropriate
cryptographic tools, entities can estimate the benefits of collaboration and
agree on what to share in a privacy-preserving way, without having to disclose
their datasets. We focus on collaborative predictive blacklisting, i.e.,
forecasting attack sources based on one's logs and those contributed by other
organizations. We study the impact of different sharing strategies by
experimenting on a real-world dataset of two billion suspicious IP addresses
collected from Dshield over two months. We find that controlled data sharing
yields up to 105% accuracy improvement on average, while also reducing the
false positive rate.
| no_new_dataset | 0.667337 |
1503.01224 | Chunhua Shen | Peng Wang, Yuanzhouhan Cao, Chunhua Shen, Lingqiao Liu, Heng Tao Shen | Temporal Pyramid Pooling Based Convolutional Neural Networks for Action
Recognition | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Encouraged by the success of Convolutional Neural Networks (CNNs) in image
classification, recently much effort is spent on applying CNNs to video based
action recognition problems. One challenge is that video contains a varying
number of frames which is incompatible to the standard input format of CNNs.
Existing methods handle this issue either by directly sampling a fixed number
of frames or bypassing this issue by introducing a 3D convolutional layer which
conducts convolution in spatial-temporal domain.
To solve this issue, here we propose a novel network structure which allows
an arbitrary number of frames as the network input. The key of our solution is
to introduce a module consisting of an encoding layer and a temporal pyramid
pooling layer. The encoding layer maps the activation from previous layers to a
feature vector suitable for pooling while the temporal pyramid pooling layer
converts multiple frame-level activations into a fixed-length video-level
representation. In addition, we adopt a feature concatenation layer which
combines appearance information and motion information. Compared with the frame
sampling strategy, our method avoids the risk of missing any important frames.
Compared with the 3D convolutional method which requires a huge video dataset
for network training, our model can be learned on a small target dataset
because we can leverage the off-the-shelf image-level CNN for model parameter
initialization. Experiments on two challenging datasets, Hollywood2 and HMDB51,
demonstrate that our method achieves superior performance over state-of-the-art
methods while requiring much fewer training data.
| [
{
"version": "v1",
"created": "Wed, 4 Mar 2015 05:18:18 GMT"
},
{
"version": "v2",
"created": "Thu, 16 Apr 2015 12:18:46 GMT"
}
] | 2015-04-17T00:00:00 | [
[
"Wang",
"Peng",
""
],
[
"Cao",
"Yuanzhouhan",
""
],
[
"Shen",
"Chunhua",
""
],
[
"Liu",
"Lingqiao",
""
],
[
"Shen",
"Heng Tao",
""
]
] | TITLE: Temporal Pyramid Pooling Based Convolutional Neural Networks for Action
Recognition
ABSTRACT: Encouraged by the success of Convolutional Neural Networks (CNNs) in image
classification, recently much effort is spent on applying CNNs to video based
action recognition problems. One challenge is that video contains a varying
number of frames which is incompatible to the standard input format of CNNs.
Existing methods handle this issue either by directly sampling a fixed number
of frames or bypassing this issue by introducing a 3D convolutional layer which
conducts convolution in spatial-temporal domain.
To solve this issue, here we propose a novel network structure which allows
an arbitrary number of frames as the network input. The key of our solution is
to introduce a module consisting of an encoding layer and a temporal pyramid
pooling layer. The encoding layer maps the activation from previous layers to a
feature vector suitable for pooling while the temporal pyramid pooling layer
converts multiple frame-level activations into a fixed-length video-level
representation. In addition, we adopt a feature concatenation layer which
combines appearance information and motion information. Compared with the frame
sampling strategy, our method avoids the risk of missing any important frames.
Compared with the 3D convolutional method which requires a huge video dataset
for network training, our model can be learned on a small target dataset
because we can leverage the off-the-shelf image-level CNN for model parameter
initialization. Experiments on two challenging datasets, Hollywood2 and HMDB51,
demonstrate that our method achieves superior performance over state-of-the-art
methods while requiring much fewer training data.
| no_new_dataset | 0.711625 |
1504.03641 | Sergey Zagoruyko | Sergey Zagoruyko and Nikos Komodakis | Learning to Compare Image Patches via Convolutional Neural Networks | CVPR 2015 | null | null | null | cs.CV cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we show how to learn directly from image data (i.e., without
resorting to manually-designed features) a general similarity function for
comparing image patches, which is a task of fundamental importance for many
computer vision problems. To encode such a function, we opt for a CNN-based
model that is trained to account for a wide variety of changes in image
appearance. To that end, we explore and study multiple neural network
architectures, which are specifically adapted to this task. We show that such
an approach can significantly outperform the state-of-the-art on several
problems and benchmark datasets.
| [
{
"version": "v1",
"created": "Tue, 14 Apr 2015 17:53:51 GMT"
}
] | 2015-04-17T00:00:00 | [
[
"Zagoruyko",
"Sergey",
""
],
[
"Komodakis",
"Nikos",
""
]
] | TITLE: Learning to Compare Image Patches via Convolutional Neural Networks
ABSTRACT: In this paper we show how to learn directly from image data (i.e., without
resorting to manually-designed features) a general similarity function for
comparing image patches, which is a task of fundamental importance for many
computer vision problems. To encode such a function, we opt for a CNN-based
model that is trained to account for a wide variety of changes in image
appearance. To that end, we explore and study multiple neural network
architectures, which are specifically adapted to this task. We show that such
an approach can significantly outperform the state-of-the-art on several
problems and benchmark datasets.
| no_new_dataset | 0.710804 |
1504.04054 | Xin Yuan | Yunchen Pu, Xin Yuan, Lawrence Carin | A Generative Model for Deep Convolutional Learning | 3 pages, 1 figure, ICLR workshop | null | null | null | stat.ML cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A generative model is developed for deep (multi-layered) convolutional
dictionary learning. A novel probabilistic pooling operation is integrated into
the deep model, yielding efficient bottom-up (pretraining) and top-down
(refinement) probabilistic learning. Experimental results demonstrate powerful
capabilities of the model to learn multi-layer features from images, and
excellent classification results are obtained on the MNIST and Caltech 101
datasets.
| [
{
"version": "v1",
"created": "Wed, 15 Apr 2015 21:31:58 GMT"
}
] | 2015-04-17T00:00:00 | [
[
"Pu",
"Yunchen",
""
],
[
"Yuan",
"Xin",
""
],
[
"Carin",
"Lawrence",
""
]
] | TITLE: A Generative Model for Deep Convolutional Learning
ABSTRACT: A generative model is developed for deep (multi-layered) convolutional
dictionary learning. A novel probabilistic pooling operation is integrated into
the deep model, yielding efficient bottom-up (pretraining) and top-down
(refinement) probabilistic learning. Experimental results demonstrate powerful
capabilities of the model to learn multi-layer features from images, and
excellent classification results are obtained on the MNIST and Caltech 101
datasets.
| no_new_dataset | 0.709648 |
1504.04208 | Andrea Scharnhorst | Rob Koopman, Shenghui Wang, Andrea Scharnhorst | Contextualization of topics - browsing through terms, authors, journals
and cluster allocations | proceedings of the ISSI 2015 conference (accepted) | null | null | null | cs.DL cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper builds on an innovative Information Retrieval tool, Ariadne. The
tool has been developed as an interactive network visualization and browsing
tool for large-scale bibliographic databases. It basically allows to gain
insights into a topic by contextualizing a search query (Koopman et al., 2015).
In this paper, we apply the Ariadne tool to a far smaller dataset of 111,616
documents in astronomy and astrophysics. Labeled as the Berlin dataset, this
data have been used by several research teams to apply and later compare
different clustering algorithms. The quest for this team effort is how to
delineate topics. This paper contributes to this challenge in two different
ways. First, we produce one of the different cluster solution and second, we
use Ariadne (the method behind it, and the interface - called LittleAriadne) to
display cluster solutions of the different group members. By providing a tool
that allows the visual inspection of the similarity of article clusters
produced by different algorithms, we present a complementary approach to other
possible means of comparison. More particular, we discuss how we can - with
LittleAriadne - browse through the network of topical terms, authors, journals
and cluster solutions in the Berlin dataset and compare cluster solutions as
well as see their context.
| [
{
"version": "v1",
"created": "Thu, 16 Apr 2015 12:38:10 GMT"
}
] | 2015-04-17T00:00:00 | [
[
"Koopman",
"Rob",
""
],
[
"Wang",
"Shenghui",
""
],
[
"Scharnhorst",
"Andrea",
""
]
] | TITLE: Contextualization of topics - browsing through terms, authors, journals
and cluster allocations
ABSTRACT: This paper builds on an innovative Information Retrieval tool, Ariadne. The
tool has been developed as an interactive network visualization and browsing
tool for large-scale bibliographic databases. It basically allows to gain
insights into a topic by contextualizing a search query (Koopman et al., 2015).
In this paper, we apply the Ariadne tool to a far smaller dataset of 111,616
documents in astronomy and astrophysics. Labeled as the Berlin dataset, this
data have been used by several research teams to apply and later compare
different clustering algorithms. The quest for this team effort is how to
delineate topics. This paper contributes to this challenge in two different
ways. First, we produce one of the different cluster solution and second, we
use Ariadne (the method behind it, and the interface - called LittleAriadne) to
display cluster solutions of the different group members. By providing a tool
that allows the visual inspection of the similarity of article clusters
produced by different algorithms, we present a complementary approach to other
possible means of comparison. More particular, we discuss how we can - with
LittleAriadne - browse through the network of topical terms, authors, journals
and cluster solutions in the Berlin dataset and compare cluster solutions as
well as see their context.
| no_new_dataset | 0.68491 |
1406.5670 | Zhirong Wu | Zhirong Wu, Shuran Song, Aditya Khosla, Fisher Yu, Linguang Zhang,
Xiaoou Tang, Jianxiong Xiao | 3D ShapeNets: A Deep Representation for Volumetric Shapes | to be appeared in CVPR 2015 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | 3D shape is a crucial but heavily underutilized cue in today's computer
vision systems, mostly due to the lack of a good generic shape representation.
With the recent availability of inexpensive 2.5D depth sensors (e.g. Microsoft
Kinect), it is becoming increasingly important to have a powerful 3D shape
representation in the loop. Apart from category recognition, recovering full 3D
shapes from view-based 2.5D depth maps is also a critical part of visual
understanding. To this end, we propose to represent a geometric 3D shape as a
probability distribution of binary variables on a 3D voxel grid, using a
Convolutional Deep Belief Network. Our model, 3D ShapeNets, learns the
distribution of complex 3D shapes across different object categories and
arbitrary poses from raw CAD data, and discovers hierarchical compositional
part representations automatically. It naturally supports joint object
recognition and shape completion from 2.5D depth maps, and it enables active
object recognition through view planning. To train our 3D deep learning model,
we construct ModelNet -- a large-scale 3D CAD model dataset. Extensive
experiments show that our 3D deep representation enables significant
performance improvement over the-state-of-the-arts in a variety of tasks.
| [
{
"version": "v1",
"created": "Sun, 22 Jun 2014 03:31:52 GMT"
},
{
"version": "v2",
"created": "Mon, 1 Sep 2014 04:59:49 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Apr 2015 16:46:05 GMT"
}
] | 2015-04-16T00:00:00 | [
[
"Wu",
"Zhirong",
""
],
[
"Song",
"Shuran",
""
],
[
"Khosla",
"Aditya",
""
],
[
"Yu",
"Fisher",
""
],
[
"Zhang",
"Linguang",
""
],
[
"Tang",
"Xiaoou",
""
],
[
"Xiao",
"Jianxiong",
""
]
] | TITLE: 3D ShapeNets: A Deep Representation for Volumetric Shapes
ABSTRACT: 3D shape is a crucial but heavily underutilized cue in today's computer
vision systems, mostly due to the lack of a good generic shape representation.
With the recent availability of inexpensive 2.5D depth sensors (e.g. Microsoft
Kinect), it is becoming increasingly important to have a powerful 3D shape
representation in the loop. Apart from category recognition, recovering full 3D
shapes from view-based 2.5D depth maps is also a critical part of visual
understanding. To this end, we propose to represent a geometric 3D shape as a
probability distribution of binary variables on a 3D voxel grid, using a
Convolutional Deep Belief Network. Our model, 3D ShapeNets, learns the
distribution of complex 3D shapes across different object categories and
arbitrary poses from raw CAD data, and discovers hierarchical compositional
part representations automatically. It naturally supports joint object
recognition and shape completion from 2.5D depth maps, and it enables active
object recognition through view planning. To train our 3D deep learning model,
we construct ModelNet -- a large-scale 3D CAD model dataset. Extensive
experiments show that our 3D deep representation enables significant
performance improvement over the-state-of-the-arts in a variety of tasks.
| new_dataset | 0.690976 |
1410.6910 | Thomas Kreuz | Thomas Kreuz and Mario Mulansky and Nebojsa Bozanic | SPIKY: A graphical user interface for monitoring spike train synchrony | 15 pages, 7 figures | null | null | null | physics.data-an cs.MS cs.SE physics.bio-ph physics.med-ph q-bio.NC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Techniques for recording large-scale neuronal spiking activity are developing
very fast. This leads to an increasing demand for algorithms capable of
analyzing large amounts of experimental spike train data. One of the most
crucial and demanding tasks is the identification of similarity patterns with a
very high temporal resolution and across different spatial scales. To address
this task, in recent years three time-resolved measures of spike train
synchrony have been proposed, the ISI-distance, the SPIKE-distance, and event
synchronization. The Matlab source codes for calculating and visualizing these
measures have been made publicly available. However, due to the many different
possible representations of the results the use of these codes is rather
complicated and their application requires some basic knowledge of Matlab. Thus
it became desirable to provide a more user-friendly and interactive interface.
Here we address this need and present SPIKY, a graphical user interface which
facilitates the application of time-resolved measures of spike train synchrony
to both simulated and real data. SPIKY includes implementations of the
ISI-distance, the SPIKE-distance and SPIKE-synchronization (an improved and
simplified extension of event synchronization) which have been optimized with
respect to computation speed and memory demand. It also comprises a spike train
generator and an event detector which makes it capable of analyzing continuous
data. Finally, the SPIKY package includes additional complementary programs
aimed at the analysis of large numbers of datasets and the estimation of
significance levels.
| [
{
"version": "v1",
"created": "Sat, 25 Oct 2014 11:02:26 GMT"
},
{
"version": "v2",
"created": "Sat, 24 Jan 2015 22:29:28 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Apr 2015 16:40:11 GMT"
}
] | 2015-04-16T00:00:00 | [
[
"Kreuz",
"Thomas",
""
],
[
"Mulansky",
"Mario",
""
],
[
"Bozanic",
"Nebojsa",
""
]
] | TITLE: SPIKY: A graphical user interface for monitoring spike train synchrony
ABSTRACT: Techniques for recording large-scale neuronal spiking activity are developing
very fast. This leads to an increasing demand for algorithms capable of
analyzing large amounts of experimental spike train data. One of the most
crucial and demanding tasks is the identification of similarity patterns with a
very high temporal resolution and across different spatial scales. To address
this task, in recent years three time-resolved measures of spike train
synchrony have been proposed, the ISI-distance, the SPIKE-distance, and event
synchronization. The Matlab source codes for calculating and visualizing these
measures have been made publicly available. However, due to the many different
possible representations of the results the use of these codes is rather
complicated and their application requires some basic knowledge of Matlab. Thus
it became desirable to provide a more user-friendly and interactive interface.
Here we address this need and present SPIKY, a graphical user interface which
facilitates the application of time-resolved measures of spike train synchrony
to both simulated and real data. SPIKY includes implementations of the
ISI-distance, the SPIKE-distance and SPIKE-synchronization (an improved and
simplified extension of event synchronization) which have been optimized with
respect to computation speed and memory demand. It also comprises a spike train
generator and an event detector which makes it capable of analyzing continuous
data. Finally, the SPIKY package includes additional complementary programs
aimed at the analysis of large numbers of datasets and the estimation of
significance levels.
| no_new_dataset | 0.707777 |
1412.6596 | Scott Reed | Scott Reed, Honglak Lee, Dragomir Anguelov, Christian Szegedy, Dumitru
Erhan, Andrew Rabinovich | Training Deep Neural Networks on Noisy Labels with Bootstrapping | null | null | null | null | cs.CV cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Current state-of-the-art deep learning systems for visual object recognition
and detection use purely supervised training with regularization such as
dropout to avoid overfitting. The performance depends critically on the amount
of labeled examples, and in current practice the labels are assumed to be
unambiguous and accurate. However, this assumption often does not hold; e.g. in
recognition, class labels may be missing; in detection, objects in the image
may not be localized; and in general, the labeling may be subjective. In this
work we propose a generic way to handle noisy and incomplete labeling by
augmenting the prediction objective with a notion of consistency. We consider a
prediction consistent if the same prediction is made given similar percepts,
where the notion of similarity is between deep network features computed from
the input data. In experiments we demonstrate that our approach yields
substantial robustness to label noise on several datasets. On MNIST handwritten
digits, we show that our model is robust to label corruption. On the Toronto
Face Database, we show that our model handles well the case of subjective
labels in emotion recognition, achieving state-of-the- art results, and can
also benefit from unlabeled face images with no modification to our method. On
the ILSVRC2014 detection challenge data, we show that our approach extends to
very deep networks, high resolution images and structured outputs, and results
in improved scalable detection.
| [
{
"version": "v1",
"created": "Sat, 20 Dec 2014 04:11:33 GMT"
},
{
"version": "v2",
"created": "Sat, 7 Feb 2015 22:30:39 GMT"
},
{
"version": "v3",
"created": "Wed, 15 Apr 2015 19:48:37 GMT"
}
] | 2015-04-16T00:00:00 | [
[
"Reed",
"Scott",
""
],
[
"Lee",
"Honglak",
""
],
[
"Anguelov",
"Dragomir",
""
],
[
"Szegedy",
"Christian",
""
],
[
"Erhan",
"Dumitru",
""
],
[
"Rabinovich",
"Andrew",
""
]
] | TITLE: Training Deep Neural Networks on Noisy Labels with Bootstrapping
ABSTRACT: Current state-of-the-art deep learning systems for visual object recognition
and detection use purely supervised training with regularization such as
dropout to avoid overfitting. The performance depends critically on the amount
of labeled examples, and in current practice the labels are assumed to be
unambiguous and accurate. However, this assumption often does not hold; e.g. in
recognition, class labels may be missing; in detection, objects in the image
may not be localized; and in general, the labeling may be subjective. In this
work we propose a generic way to handle noisy and incomplete labeling by
augmenting the prediction objective with a notion of consistency. We consider a
prediction consistent if the same prediction is made given similar percepts,
where the notion of similarity is between deep network features computed from
the input data. In experiments we demonstrate that our approach yields
substantial robustness to label noise on several datasets. On MNIST handwritten
digits, we show that our model is robust to label corruption. On the Toronto
Face Database, we show that our model handles well the case of subjective
labels in emotion recognition, achieving state-of-the- art results, and can
also benefit from unlabeled face images with no modification to our method. On
the ILSVRC2014 detection challenge data, we show that our approach extends to
very deep networks, high resolution images and structured outputs, and results
in improved scalable detection.
| no_new_dataset | 0.708629 |
1504.03961 | Tao Chen | Tao Chen, Rami Bahsoon and Xin Yao | Online QoS Modeling in the Cloud: A Hybrid and Adaptive Multi-Learners
Approach | In the proceeding of the 7th IEEE/ACM International Conference on
Utility and Cloud Computing (UCC), London, UK, 2014 | null | 10.1109/UCC.2014.42 | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Given the on-demand nature of cloud computing, managing cloud-based services
requires accurate modeling for the correlation between their Quality of Service
(QoS) and cloud configurations/resources. The resulted models need to cope with
the dynamic fluctuation of QoS sensitivity and interference. However, existing
QoS modeling in the cloud are limited in terms of both accuracy and
applicability due to their static and semi- dynamic nature. In this paper, we
present a fully dynamic multi- learners approach for automated and online QoS
modeling in the cloud. We contribute to a hybrid learners solution, which
improves accuracy while keeping model complexity adequate. To determine the
inputs of QoS model at runtime, we partition the inputs space into two
sub-spaces, each of which applies different symmetric uncertainty based
selection techniques, and we then combine the sub-spaces results. The learners
are also adaptive; they simultaneously allow several machine learning
algorithms to model QoS function and dynamically select the best model for
prediction on the fly. We experimentally evaluate our models using RUBiS
benchmark and realistic FIFA 98 workload. The results show that our
multi-learners approach is more accurate and effective in contrast to the other
state-of-the-art approaches.
| [
{
"version": "v1",
"created": "Wed, 15 Apr 2015 16:35:40 GMT"
}
] | 2015-04-16T00:00:00 | [
[
"Chen",
"Tao",
""
],
[
"Bahsoon",
"Rami",
""
],
[
"Yao",
"Xin",
""
]
] | TITLE: Online QoS Modeling in the Cloud: A Hybrid and Adaptive Multi-Learners
Approach
ABSTRACT: Given the on-demand nature of cloud computing, managing cloud-based services
requires accurate modeling for the correlation between their Quality of Service
(QoS) and cloud configurations/resources. The resulted models need to cope with
the dynamic fluctuation of QoS sensitivity and interference. However, existing
QoS modeling in the cloud are limited in terms of both accuracy and
applicability due to their static and semi- dynamic nature. In this paper, we
present a fully dynamic multi- learners approach for automated and online QoS
modeling in the cloud. We contribute to a hybrid learners solution, which
improves accuracy while keeping model complexity adequate. To determine the
inputs of QoS model at runtime, we partition the inputs space into two
sub-spaces, each of which applies different symmetric uncertainty based
selection techniques, and we then combine the sub-spaces results. The learners
are also adaptive; they simultaneously allow several machine learning
algorithms to model QoS function and dynamically select the best model for
prediction on the fly. We experimentally evaluate our models using RUBiS
benchmark and realistic FIFA 98 workload. The results show that our
multi-learners approach is more accurate and effective in contrast to the other
state-of-the-art approaches.
| no_new_dataset | 0.709837 |
1404.3708 | Yuxiao Dong | Yuxiao Dong, Jie Tang, Nitesh Chawla, Tiancheng Lou, Yang Yang, Bai
Wang | Inferring Social Status and Rich Club Effects in Enterprise
Communication Networks | 13 pages, 4 figures | PLoS ONE 10(3): e0119446. 2015 | 10.1371/journal.pone.0119446 | null | cs.SI cs.AI physics.soc-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Social status, defined as the relative rank or position that an individual
holds in a social hierarchy, is known to be among the most important motivating
forces in social behaviors. In this paper, we consider the notion of status
from the perspective of a position or title held by a person in an enterprise.
We study the intersection of social status and social networks in an
enterprise. We study whether enterprise communication logs can help reveal how
social interactions and individual status manifest themselves in social
networks. To that end, we use two enterprise datasets with three communication
channels --- voice call, short message, and email --- to demonstrate the
social-behavioral differences among individuals with different status. We have
several interesting findings and based on these findings we also develop a
model to predict social status. On the individual level, high-status
individuals are more likely to be spanned as structural holes by linking to
people in parts of the enterprise networks that are otherwise not well
connected to one another. On the community level, the principle of homophily,
social balance and clique theory generally indicate a "rich club" maintained by
high-status individuals, in the sense that this community is much more
connected, balanced and dense. Our model can predict social status of
individuals with 93% accuracy.
| [
{
"version": "v1",
"created": "Mon, 14 Apr 2014 19:32:08 GMT"
},
{
"version": "v2",
"created": "Tue, 15 Apr 2014 01:25:26 GMT"
},
{
"version": "v3",
"created": "Wed, 16 Apr 2014 20:12:21 GMT"
},
{
"version": "v4",
"created": "Tue, 14 Apr 2015 19:44:47 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Dong",
"Yuxiao",
""
],
[
"Tang",
"Jie",
""
],
[
"Chawla",
"Nitesh",
""
],
[
"Lou",
"Tiancheng",
""
],
[
"Yang",
"Yang",
""
],
[
"Wang",
"Bai",
""
]
] | TITLE: Inferring Social Status and Rich Club Effects in Enterprise
Communication Networks
ABSTRACT: Social status, defined as the relative rank or position that an individual
holds in a social hierarchy, is known to be among the most important motivating
forces in social behaviors. In this paper, we consider the notion of status
from the perspective of a position or title held by a person in an enterprise.
We study the intersection of social status and social networks in an
enterprise. We study whether enterprise communication logs can help reveal how
social interactions and individual status manifest themselves in social
networks. To that end, we use two enterprise datasets with three communication
channels --- voice call, short message, and email --- to demonstrate the
social-behavioral differences among individuals with different status. We have
several interesting findings and based on these findings we also develop a
model to predict social status. On the individual level, high-status
individuals are more likely to be spanned as structural holes by linking to
people in parts of the enterprise networks that are otherwise not well
connected to one another. On the community level, the principle of homophily,
social balance and clique theory generally indicate a "rich club" maintained by
high-status individuals, in the sense that this community is much more
connected, balanced and dense. Our model can predict social status of
individuals with 93% accuracy.
| no_new_dataset | 0.709202 |
1411.4894 | Ayan Chakrabarti | Ayan Chakrabarti, Ying Xiong, Steven J. Gortler, Todd Zickler | Low-level Vision by Consensus in a Spatial Hierarchy of Regions | Accepted to CVPR 2015. Project page:
http://www.ttic.edu/chakrabarti/consensus/ | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a multi-scale framework for low-level vision, where the goal is
estimating physical scene values from image data---such as depth from stereo
image pairs. The framework uses a dense, overlapping set of image regions at
multiple scales and a "local model," such as a slanted-plane model for stereo
disparity, that is expected to be valid piecewise across the visual field.
Estimation is cast as optimization over a dichotomous mixture of variables,
simultaneously determining which regions are inliers with respect to the local
model (binary variables) and the correct co-ordinates in the local model space
for each inlying region (continuous variables). When the regions are organized
into a multi-scale hierarchy, optimization can occur in an efficient and
parallel architecture, where distributed computational units iteratively
perform calculations and share information through sparse connections between
parents and children. The framework performs well on a standard benchmark for
binocular stereo, and it produces a distributional scene representation that is
appropriate for combining with higher-level reasoning and other low-level cues.
| [
{
"version": "v1",
"created": "Tue, 18 Nov 2014 16:23:06 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Apr 2015 15:57:52 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Chakrabarti",
"Ayan",
""
],
[
"Xiong",
"Ying",
""
],
[
"Gortler",
"Steven J.",
""
],
[
"Zickler",
"Todd",
""
]
] | TITLE: Low-level Vision by Consensus in a Spatial Hierarchy of Regions
ABSTRACT: We introduce a multi-scale framework for low-level vision, where the goal is
estimating physical scene values from image data---such as depth from stereo
image pairs. The framework uses a dense, overlapping set of image regions at
multiple scales and a "local model," such as a slanted-plane model for stereo
disparity, that is expected to be valid piecewise across the visual field.
Estimation is cast as optimization over a dichotomous mixture of variables,
simultaneously determining which regions are inliers with respect to the local
model (binary variables) and the correct co-ordinates in the local model space
for each inlying region (continuous variables). When the regions are organized
into a multi-scale hierarchy, optimization can occur in an efficient and
parallel architecture, where distributed computational units iteratively
perform calculations and share information through sparse connections between
parents and children. The framework performs well on a standard benchmark for
binocular stereo, and it produces a distributional scene representation that is
appropriate for combining with higher-level reasoning and other low-level cues.
| no_new_dataset | 0.711443 |
1412.0623 | Sean Bell | Sean Bell and Paul Upchurch and Noah Snavely and Kavita Bala | Material Recognition in the Wild with the Materials in Context Database | CVPR 2015. Sean Bell and Paul Upchurch contributed equally | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recognizing materials in real-world images is a challenging task. Real-world
materials have rich surface texture, geometry, lighting conditions, and
clutter, which combine to make the problem particularly difficult. In this
paper, we introduce a new, large-scale, open dataset of materials in the wild,
the Materials in Context Database (MINC), and combine this dataset with deep
learning to achieve material recognition and segmentation of images in the
wild.
MINC is an order of magnitude larger than previous material databases, while
being more diverse and well-sampled across its 23 categories. Using MINC, we
train convolutional neural networks (CNNs) for two tasks: classifying materials
from patches, and simultaneous material recognition and segmentation in full
images. For patch-based classification on MINC we found that the best
performing CNN architectures can achieve 85.2% mean class accuracy. We convert
these trained CNN classifiers into an efficient fully convolutional framework
combined with a fully connected conditional random field (CRF) to predict the
material at every pixel in an image, achieving 73.1% mean class accuracy. Our
experiments demonstrate that having a large, well-sampled dataset such as MINC
is crucial for real-world material recognition and segmentation.
| [
{
"version": "v1",
"created": "Mon, 1 Dec 2014 20:11:44 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Apr 2015 05:29:32 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Bell",
"Sean",
""
],
[
"Upchurch",
"Paul",
""
],
[
"Snavely",
"Noah",
""
],
[
"Bala",
"Kavita",
""
]
] | TITLE: Material Recognition in the Wild with the Materials in Context Database
ABSTRACT: Recognizing materials in real-world images is a challenging task. Real-world
materials have rich surface texture, geometry, lighting conditions, and
clutter, which combine to make the problem particularly difficult. In this
paper, we introduce a new, large-scale, open dataset of materials in the wild,
the Materials in Context Database (MINC), and combine this dataset with deep
learning to achieve material recognition and segmentation of images in the
wild.
MINC is an order of magnitude larger than previous material databases, while
being more diverse and well-sampled across its 23 categories. Using MINC, we
train convolutional neural networks (CNNs) for two tasks: classifying materials
from patches, and simultaneous material recognition and segmentation in full
images. For patch-based classification on MINC we found that the best
performing CNN architectures can achieve 85.2% mean class accuracy. We convert
these trained CNN classifiers into an efficient fully convolutional framework
combined with a fully connected conditional random field (CRF) to predict the
material at every pixel in an image, achieving 73.1% mean class accuracy. Our
experiments demonstrate that having a large, well-sampled dataset such as MINC
is crucial for real-world material recognition and segmentation.
| new_dataset | 0.716219 |
1412.2306 | Andrej Karpathy | Andrej Karpathy, Li Fei-Fei | Deep Visual-Semantic Alignments for Generating Image Descriptions | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a model that generates natural language descriptions of images and
their regions. Our approach leverages datasets of images and their sentence
descriptions to learn about the inter-modal correspondences between language
and visual data. Our alignment model is based on a novel combination of
Convolutional Neural Networks over image regions, bidirectional Recurrent
Neural Networks over sentences, and a structured objective that aligns the two
modalities through a multimodal embedding. We then describe a Multimodal
Recurrent Neural Network architecture that uses the inferred alignments to
learn to generate novel descriptions of image regions. We demonstrate that our
alignment model produces state of the art results in retrieval experiments on
Flickr8K, Flickr30K and MSCOCO datasets. We then show that the generated
descriptions significantly outperform retrieval baselines on both full images
and on a new dataset of region-level annotations.
| [
{
"version": "v1",
"created": "Sun, 7 Dec 2014 02:36:07 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Apr 2015 05:02:53 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Karpathy",
"Andrej",
""
],
[
"Fei-Fei",
"Li",
""
]
] | TITLE: Deep Visual-Semantic Alignments for Generating Image Descriptions
ABSTRACT: We present a model that generates natural language descriptions of images and
their regions. Our approach leverages datasets of images and their sentence
descriptions to learn about the inter-modal correspondences between language
and visual data. Our alignment model is based on a novel combination of
Convolutional Neural Networks over image regions, bidirectional Recurrent
Neural Networks over sentences, and a structured objective that aligns the two
modalities through a multimodal embedding. We then describe a Multimodal
Recurrent Neural Network architecture that uses the inferred alignments to
learn to generate novel descriptions of image regions. We demonstrate that our
alignment model produces state of the art results in retrieval experiments on
Flickr8K, Flickr30K and MSCOCO datasets. We then show that the generated
descriptions significantly outperform retrieval baselines on both full images
and on a new dataset of region-level annotations.
| new_dataset | 0.711462 |
1412.3709 | Abel Gonzalez-Garcia | Abel Gonzalez-Garcia, Alexander Vezhnevets, Vittorio Ferrari | An active search strategy for efficient object class detection | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Object class detectors typically apply a window classifier to all the windows
in a large set, either in a sliding window manner or using object proposals. In
this paper, we develop an active search strategy that sequentially chooses the
next window to evaluate based on all the information gathered before. This
results in a substantial reduction in the number of classifier evaluations and
in a more elegant approach in general. Our search strategy is guided by two
forces. First, we exploit context as the statistical relation between the
appearance of a window and its location relative to the object, as observed in
the training set. This enables to jump across distant regions in the image
(e.g. observing a sky region suggests that cars might be far below) and is done
efficiently in a Random Forest framework. Second, we exploit the score of the
classifier to attract the search to promising areas surrounding a highly scored
window, and to keep away from areas near low scored ones. Our search strategy
can be applied on top of any classifier as it treats it as a black-box. In
experiments with R-CNN on the challenging SUN2012 dataset, our method matches
the detection accuracy of evaluating all windows independently, while
evaluating 9x fewer windows.
| [
{
"version": "v1",
"created": "Thu, 11 Dec 2014 16:23:38 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Apr 2015 11:29:51 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Gonzalez-Garcia",
"Abel",
""
],
[
"Vezhnevets",
"Alexander",
""
],
[
"Ferrari",
"Vittorio",
""
]
] | TITLE: An active search strategy for efficient object class detection
ABSTRACT: Object class detectors typically apply a window classifier to all the windows
in a large set, either in a sliding window manner or using object proposals. In
this paper, we develop an active search strategy that sequentially chooses the
next window to evaluate based on all the information gathered before. This
results in a substantial reduction in the number of classifier evaluations and
in a more elegant approach in general. Our search strategy is guided by two
forces. First, we exploit context as the statistical relation between the
appearance of a window and its location relative to the object, as observed in
the training set. This enables to jump across distant regions in the image
(e.g. observing a sky region suggests that cars might be far below) and is done
efficiently in a Random Forest framework. Second, we exploit the score of the
classifier to attract the search to promising areas surrounding a highly scored
window, and to keep away from areas near low scored ones. Our search strategy
can be applied on top of any classifier as it treats it as a black-box. In
experiments with R-CNN on the challenging SUN2012 dataset, our method matches
the detection accuracy of evaluating all windows independently, while
evaluating 9x fewer windows.
| no_new_dataset | 0.711638 |
1412.6599 | Kevin Bache | Kevin Bache, Dennis DeCoste, Padhraic Smyth | Hot Swapping for Online Adaptation of Optimization Hyperparameters | Submission to ICLR 2015 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We describe a general framework for online adaptation of optimization
hyperparameters by `hot swapping' their values during learning. We investigate
this approach in the context of adaptive learning rate selection using an
explore-exploit strategy from the multi-armed bandit literature. Experiments on
a benchmark neural network show that the hot swapping approach leads to
consistently better solutions compared to well-known alternatives such as
AdaDelta and stochastic gradient with exhaustive hyperparameter search.
| [
{
"version": "v1",
"created": "Sat, 20 Dec 2014 04:36:28 GMT"
},
{
"version": "v2",
"created": "Sat, 28 Feb 2015 05:18:18 GMT"
},
{
"version": "v3",
"created": "Mon, 13 Apr 2015 23:28:01 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Bache",
"Kevin",
""
],
[
"DeCoste",
"Dennis",
""
],
[
"Smyth",
"Padhraic",
""
]
] | TITLE: Hot Swapping for Online Adaptation of Optimization Hyperparameters
ABSTRACT: We describe a general framework for online adaptation of optimization
hyperparameters by `hot swapping' their values during learning. We investigate
this approach in the context of adaptive learning rate selection using an
explore-exploit strategy from the multi-armed bandit literature. Experiments on
a benchmark neural network show that the hot swapping approach leads to
consistently better solutions compared to well-known alternatives such as
AdaDelta and stochastic gradient with exhaustive hyperparameter search.
| no_new_dataset | 0.71039 |
1501.00100 | Marco Gramaglia | Marco Gramaglia and Marco Fiore | On the anonymizability of mobile traffic datasets | null | null | null | null | cs.CY cs.CR | http://creativecommons.org/licenses/by-nc-sa/3.0/ | Preserving user privacy is paramount when it comes to publicly disclosed
datasets that contain fine-grained data about large populations. The problem is
especially critical in the case of mobile traffic datasets collected by
cellular operators, as they feature elevate subscriber trajectory uniqueness
and they are resistant to anonymization through spatiotemporal generalization.
In this work, we investigate the $k$-anonymizability of trajectories in two
large-scale mobile traffic datasets, by means of a novel dedicated measure. Our
results are in agreement with those of previous analyses, however they also
provide additional insights on the reasons behind the poor anonimizability of
mobile traffic datasets. As such, our study is a step forward in the direction
of a more robust dataset anonymization.
| [
{
"version": "v1",
"created": "Wed, 31 Dec 2014 09:53:31 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Apr 2015 09:35:41 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Gramaglia",
"Marco",
""
],
[
"Fiore",
"Marco",
""
]
] | TITLE: On the anonymizability of mobile traffic datasets
ABSTRACT: Preserving user privacy is paramount when it comes to publicly disclosed
datasets that contain fine-grained data about large populations. The problem is
especially critical in the case of mobile traffic datasets collected by
cellular operators, as they feature elevate subscriber trajectory uniqueness
and they are resistant to anonymization through spatiotemporal generalization.
In this work, we investigate the $k$-anonymizability of trajectories in two
large-scale mobile traffic datasets, by means of a novel dedicated measure. Our
results are in agreement with those of previous analyses, however they also
provide additional insights on the reasons behind the poor anonimizability of
mobile traffic datasets. As such, our study is a step forward in the direction
of a more robust dataset anonymization.
| no_new_dataset | 0.711857 |
1502.01916 | Daniel Lemire | Wayne Xin Zhao, Xudong Zhang, Daniel Lemire, Dongdong Shan, Jian-Yun
Nie, Hongfei Yan, Ji-Rong Wen | A General SIMD-based Approach to Accelerating Compression Algorithms | null | ACM Trans. Inf. Syst. 33, 3, Article 15 (March 2015) | 10.1145/2735629 | null | cs.IR | http://creativecommons.org/licenses/by/3.0/ | Compression algorithms are important for data oriented tasks, especially in
the era of Big Data. Modern processors equipped with powerful SIMD instruction
sets, provide us an opportunity for achieving better compression performance.
Previous research has shown that SIMD-based optimizations can multiply decoding
speeds. Following these pioneering studies, we propose a general approach to
accelerate compression algorithms. By instantiating the approach, we have
developed several novel integer compression algorithms, called Group-Simple,
Group-Scheme, Group-AFOR, and Group-PFD, and implemented their corresponding
vectorized versions. We evaluate the proposed algorithms on two public TREC
datasets, a Wikipedia dataset and a Twitter dataset. With competitive
compression ratios and encoding speeds, our SIMD-based algorithms outperform
state-of-the-art non-vectorized algorithms with respect to decoding speeds.
| [
{
"version": "v1",
"created": "Fri, 6 Feb 2015 15:11:47 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Zhao",
"Wayne Xin",
""
],
[
"Zhang",
"Xudong",
""
],
[
"Lemire",
"Daniel",
""
],
[
"Shan",
"Dongdong",
""
],
[
"Nie",
"Jian-Yun",
""
],
[
"Yan",
"Hongfei",
""
],
[
"Wen",
"Ji-Rong",
""
]
] | TITLE: A General SIMD-based Approach to Accelerating Compression Algorithms
ABSTRACT: Compression algorithms are important for data oriented tasks, especially in
the era of Big Data. Modern processors equipped with powerful SIMD instruction
sets, provide us an opportunity for achieving better compression performance.
Previous research has shown that SIMD-based optimizations can multiply decoding
speeds. Following these pioneering studies, we propose a general approach to
accelerate compression algorithms. By instantiating the approach, we have
developed several novel integer compression algorithms, called Group-Simple,
Group-Scheme, Group-AFOR, and Group-PFD, and implemented their corresponding
vectorized versions. We evaluate the proposed algorithms on two public TREC
datasets, a Wikipedia dataset and a Twitter dataset. With competitive
compression ratios and encoding speeds, our SIMD-based algorithms outperform
state-of-the-art non-vectorized algorithms with respect to decoding speeds.
| no_new_dataset | 0.704402 |
1504.03154 | Carlo Ciliberto | Giulia Pasquale, Carlo Ciliberto, Francesca Odone, Lorenzo Rosasco and
Lorenzo Natale | Real-world Object Recognition with Off-the-shelf Deep Conv Nets: How
Many Objects can iCub Learn? | 18 pages, 9 figures, 3 tables | null | null | null | cs.RO cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The ability to visually recognize objects is a fundamental skill for robotics
systems. Indeed, a large variety of tasks involving manipulation, navigation or
interaction with other agents, deeply depends on the accurate understanding of
the visual scene. Yet, at the time being, robots are lacking good visual
perceptual systems, which often become the main bottleneck preventing the use
of autonomous agents for real-world applications.
Lately in computer vision, systems that learn suitable visual representations
and based on multi-layer deep convolutional networks are showing remarkable
performance in tasks such as large-scale visual recognition and image
retrieval. To this regard, it is natural to ask whether such remarkable
performance would generalize also to the robotic setting.
In this paper we investigate such possibility, while taking further steps in
developing a computational vision system to be embedded on a robotic platform,
the iCub humanoid robot. In particular, we release a new dataset ({\sc
iCubWorld28}) that we use as a benchmark to address the question: {\it how many
objects can iCub recognize?} Our study is developed in a learning framework
which reflects the typical visual experience of a humanoid robot like the iCub.
Experiments shed interesting insights on the strength and weaknesses of current
computer vision approaches applied in real robotic settings.
| [
{
"version": "v1",
"created": "Mon, 13 Apr 2015 12:45:09 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Apr 2015 05:56:01 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Pasquale",
"Giulia",
""
],
[
"Ciliberto",
"Carlo",
""
],
[
"Odone",
"Francesca",
""
],
[
"Rosasco",
"Lorenzo",
""
],
[
"Natale",
"Lorenzo",
""
]
] | TITLE: Real-world Object Recognition with Off-the-shelf Deep Conv Nets: How
Many Objects can iCub Learn?
ABSTRACT: The ability to visually recognize objects is a fundamental skill for robotics
systems. Indeed, a large variety of tasks involving manipulation, navigation or
interaction with other agents, deeply depends on the accurate understanding of
the visual scene. Yet, at the time being, robots are lacking good visual
perceptual systems, which often become the main bottleneck preventing the use
of autonomous agents for real-world applications.
Lately in computer vision, systems that learn suitable visual representations
and based on multi-layer deep convolutional networks are showing remarkable
performance in tasks such as large-scale visual recognition and image
retrieval. To this regard, it is natural to ask whether such remarkable
performance would generalize also to the robotic setting.
In this paper we investigate such possibility, while taking further steps in
developing a computational vision system to be embedded on a robotic platform,
the iCub humanoid robot. In particular, we release a new dataset ({\sc
iCubWorld28}) that we use as a benchmark to address the question: {\it how many
objects can iCub recognize?} Our study is developed in a learning framework
which reflects the typical visual experience of a humanoid robot like the iCub.
Experiments shed interesting insights on the strength and weaknesses of current
computer vision approaches applied in real robotic settings.
| new_dataset | 0.715182 |
1504.03415 | Darshana Wickramarachchi Mr | D. C. Wickramarachchi, B. L. Robertson, M. Reale, C. J. Price and J.
Brown | HHCART: An Oblique Decision Tree | 13 Pages, 1 Figure, 4 Tables, 1 Algorithm | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Decision trees are a popular technique in statistical data classification.
They recursively partition the feature space into disjoint sub-regions until
each sub-region becomes homogeneous with respect to a particular class. The
basic Classification and Regression Tree (CART) algorithm partitions the
feature space using axis parallel splits. When the true decision boundaries are
not aligned with the feature axes, this approach can produce a complicated
boundary structure. Oblique decision trees use oblique decision boundaries to
potentially simplify the boundary structure. The major limitation of this
approach is that the tree induction algorithm is computationally expensive. In
this article we present a new decision tree algorithm, called HHCART. The
method utilizes a series of Householder matrices to reflect the training data
at each node during the tree construction. Each reflection is based on the
directions of the eigenvectors from each classes' covariance matrix.
Considering axis parallel splits in the reflected training data provides an
efficient way of finding oblique splits in the unreflected training data.
Experimental results show that the accuracy and size of the HHCART trees are
comparable with some benchmark methods in the literature. The appealing feature
of HHCART is that it can handle both qualitative and quantitative features in
the same oblique split.
| [
{
"version": "v1",
"created": "Tue, 14 Apr 2015 04:04:00 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Wickramarachchi",
"D. C.",
""
],
[
"Robertson",
"B. L.",
""
],
[
"Reale",
"M.",
""
],
[
"Price",
"C. J.",
""
],
[
"Brown",
"J.",
""
]
] | TITLE: HHCART: An Oblique Decision Tree
ABSTRACT: Decision trees are a popular technique in statistical data classification.
They recursively partition the feature space into disjoint sub-regions until
each sub-region becomes homogeneous with respect to a particular class. The
basic Classification and Regression Tree (CART) algorithm partitions the
feature space using axis parallel splits. When the true decision boundaries are
not aligned with the feature axes, this approach can produce a complicated
boundary structure. Oblique decision trees use oblique decision boundaries to
potentially simplify the boundary structure. The major limitation of this
approach is that the tree induction algorithm is computationally expensive. In
this article we present a new decision tree algorithm, called HHCART. The
method utilizes a series of Householder matrices to reflect the training data
at each node during the tree construction. Each reflection is based on the
directions of the eigenvectors from each classes' covariance matrix.
Considering axis parallel splits in the reflected training data provides an
efficient way of finding oblique splits in the unreflected training data.
Experimental results show that the accuracy and size of the HHCART trees are
comparable with some benchmark methods in the literature. The appealing feature
of HHCART is that it can handle both qualitative and quantitative features in
the same oblique split.
| no_new_dataset | 0.710459 |
1504.03425 | Iftekhar Naim | Iftekhar Naim, M. Iftekhar Tanveer, Daniel Gildea, Mohammed (Ehsan)
Hoque | Automated Analysis and Prediction of Job Interview Performance | 14 pages, 8 figures, 6 tables | null | null | null | cs.HC cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a computational framework for automatically quantifying verbal and
nonverbal behaviors in the context of job interviews. The proposed framework is
trained by analyzing the videos of 138 interview sessions with 69
internship-seeking undergraduates at the Massachusetts Institute of Technology
(MIT). Our automated analysis includes facial expressions (e.g., smiles, head
gestures, facial tracking points), language (e.g., word counts, topic
modeling), and prosodic information (e.g., pitch, intonation, and pauses) of
the interviewees. The ground truth labels are derived by taking a weighted
average over the ratings of 9 independent judges. Our framework can
automatically predict the ratings for interview traits such as excitement,
friendliness, and engagement with correlation coefficients of 0.75 or higher,
and can quantify the relative importance of prosody, language, and facial
expressions. By analyzing the relative feature weights learned by the
regression models, our framework recommends to speak more fluently, use less
filler words, speak as "we" (vs. "I"), use more unique words, and smile more.
We also find that the students who were rated highly while answering the first
interview question were also rated highly overall (i.e., first impression
matters). Finally, our MIT Interview dataset will be made available to other
researchers to further validate and expand our findings.
| [
{
"version": "v1",
"created": "Tue, 14 Apr 2015 05:49:26 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Naim",
"Iftekhar",
"",
"Ehsan"
],
[
"Tanveer",
"M. Iftekhar",
"",
"Ehsan"
],
[
"Gildea",
"Daniel",
"",
"Ehsan"
],
[
"Mohammed",
"",
"",
"Ehsan"
],
[
"Hoque",
"",
""
]
] | TITLE: Automated Analysis and Prediction of Job Interview Performance
ABSTRACT: We present a computational framework for automatically quantifying verbal and
nonverbal behaviors in the context of job interviews. The proposed framework is
trained by analyzing the videos of 138 interview sessions with 69
internship-seeking undergraduates at the Massachusetts Institute of Technology
(MIT). Our automated analysis includes facial expressions (e.g., smiles, head
gestures, facial tracking points), language (e.g., word counts, topic
modeling), and prosodic information (e.g., pitch, intonation, and pauses) of
the interviewees. The ground truth labels are derived by taking a weighted
average over the ratings of 9 independent judges. Our framework can
automatically predict the ratings for interview traits such as excitement,
friendliness, and engagement with correlation coefficients of 0.75 or higher,
and can quantify the relative importance of prosody, language, and facial
expressions. By analyzing the relative feature weights learned by the
regression models, our framework recommends to speak more fluently, use less
filler words, speak as "we" (vs. "I"), use more unique words, and smile more.
We also find that the students who were rated highly while answering the first
interview question were also rated highly overall (i.e., first impression
matters). Finally, our MIT Interview dataset will be made available to other
researchers to further validate and expand our findings.
| new_dataset | 0.712432 |
1504.03504 | Fang Wang | Fang Wang, Le Kang, Yi Li | Sketch-based 3D Shape Retrieval using Convolutional Neural Networks | CVPR 2015 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Retrieving 3D models from 2D human sketches has received considerable
attention in the areas of graphics, image retrieval, and computer vision.
Almost always in state of the art approaches a large amount of "best views" are
computed for 3D models, with the hope that the query sketch matches one of
these 2D projections of 3D models using predefined features.
We argue that this two stage approach (view selection -- matching) is
pragmatic but also problematic because the "best views" are subjective and
ambiguous, which makes the matching inputs obscure. This imprecise nature of
matching further makes it challenging to choose features manually. Instead of
relying on the elusive concept of "best views" and the hand-crafted features,
we propose to define our views using a minimalism approach and learn features
for both sketches and views. Specifically, we drastically reduce the number of
views to only two predefined directions for the whole dataset. Then, we learn
two Siamese Convolutional Neural Networks (CNNs), one for the views and one for
the sketches. The loss function is defined on the within-domain as well as the
cross-domain similarities. Our experiments on three benchmark datasets
demonstrate that our method is significantly better than state of the art
approaches, and outperforms them in all conventional metrics.
| [
{
"version": "v1",
"created": "Tue, 14 Apr 2015 11:55:45 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Wang",
"Fang",
""
],
[
"Kang",
"Le",
""
],
[
"Li",
"Yi",
""
]
] | TITLE: Sketch-based 3D Shape Retrieval using Convolutional Neural Networks
ABSTRACT: Retrieving 3D models from 2D human sketches has received considerable
attention in the areas of graphics, image retrieval, and computer vision.
Almost always in state of the art approaches a large amount of "best views" are
computed for 3D models, with the hope that the query sketch matches one of
these 2D projections of 3D models using predefined features.
We argue that this two stage approach (view selection -- matching) is
pragmatic but also problematic because the "best views" are subjective and
ambiguous, which makes the matching inputs obscure. This imprecise nature of
matching further makes it challenging to choose features manually. Instead of
relying on the elusive concept of "best views" and the hand-crafted features,
we propose to define our views using a minimalism approach and learn features
for both sketches and views. Specifically, we drastically reduce the number of
views to only two predefined directions for the whole dataset. Then, we learn
two Siamese Convolutional Neural Networks (CNNs), one for the views and one for
the sketches. The loss function is defined on the within-domain as well as the
cross-domain similarities. Our experiments on three benchmark datasets
demonstrate that our method is significantly better than state of the art
approaches, and outperforms them in all conventional metrics.
| no_new_dataset | 0.710584 |
1504.03522 | Lukas Neumann | Luk\'a\v{s} Neumann, Ji\v{r}\'i Matas | Efficient Scene Text Localization and Recognition with Local Character
Refinement | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | An unconstrained end-to-end text localization and recognition method is
presented. The method detects initial text hypothesis in a single pass by an
efficient region-based method and subsequently refines the text hypothesis
using a more robust local text model, which deviates from the common assumption
of region-based methods that all characters are detected as connected
components.
Additionally, a novel feature based on character stroke area estimation is
introduced. The feature is efficiently computed from a region distance map, it
is invariant to scaling and rotations and allows to efficiently detect text
regions regardless of what portion of text they capture.
The method runs in real time and achieves state-of-the-art text localization
and recognition results on the ICDAR 2013 Robust Reading dataset.
| [
{
"version": "v1",
"created": "Tue, 14 Apr 2015 12:42:56 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Neumann",
"Lukáš",
""
],
[
"Matas",
"Jiří",
""
]
] | TITLE: Efficient Scene Text Localization and Recognition with Local Character
Refinement
ABSTRACT: An unconstrained end-to-end text localization and recognition method is
presented. The method detects initial text hypothesis in a single pass by an
efficient region-based method and subsequently refines the text hypothesis
using a more robust local text model, which deviates from the common assumption
of region-based methods that all characters are detected as connected
components.
Additionally, a novel feature based on character stroke area estimation is
introduced. The feature is efficiently computed from a region distance map, it
is invariant to scaling and rotations and allows to efficiently detect text
regions regardless of what portion of text they capture.
The method runs in real time and achieves state-of-the-art text localization
and recognition results on the ICDAR 2013 Robust Reading dataset.
| no_new_dataset | 0.710415 |
1504.03659 | Azad Dehghan Mr | Azad Dehghan | Temporal ordering of clinical events | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This report describes a minimalistic set of methods engineered to anchor
clinical events onto a temporal space. Specifically, we describe methods to
extract clinical events (e.g., Problems, Treatments and Tests), temporal
expressions (i.e., time, date, duration, and frequency), and temporal links
(e.g., Before, After, Overlap) between events and temporal entities. These
methods are developed and validated using high quality datasets.
| [
{
"version": "v1",
"created": "Tue, 14 Apr 2015 18:48:58 GMT"
}
] | 2015-04-15T00:00:00 | [
[
"Dehghan",
"Azad",
""
]
] | TITLE: Temporal ordering of clinical events
ABSTRACT: This report describes a minimalistic set of methods engineered to anchor
clinical events onto a temporal space. Specifically, we describe methods to
extract clinical events (e.g., Problems, Treatments and Tests), temporal
expressions (i.e., time, date, duration, and frequency), and temporal links
(e.g., Before, After, Overlap) between events and temporal entities. These
methods are developed and validated using high quality datasets.
| no_new_dataset | 0.709453 |
1409.3518 | Do-kyum Kim | Do-kyum Kim, Geoffrey M. Voelker, Lawrence K. Saul | Topic Modeling of Hierarchical Corpora | null | null | null | null | stat.ML cs.IR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the problem of topic modeling in corpora whose documents are
organized in a multi-level hierarchy. We explore a parametric approach to this
problem, assuming that the number of topics is known or can be estimated by
cross-validation. The models we consider can be viewed as special
(finite-dimensional) instances of hierarchical Dirichlet processes (HDPs). For
these models we show that there exists a simple variational approximation for
probabilistic inference. The approximation relies on a previously unexploited
inequality that handles the conditional dependence between Dirichlet latent
variables in adjacent levels of the model's hierarchy. We compare our approach
to existing implementations of nonparametric HDPs. On several benchmarks we
find that our approach is faster than Gibbs sampling and able to learn more
predictive models than existing variational methods. Finally, we demonstrate
the large-scale viability of our approach on two newly available corpora from
researchers in computer security---one with 350,000 documents and over 6,000
internal subcategories, the other with a five-level deep hierarchy.
| [
{
"version": "v1",
"created": "Thu, 11 Sep 2014 18:00:59 GMT"
},
{
"version": "v2",
"created": "Mon, 13 Apr 2015 06:29:46 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Kim",
"Do-kyum",
""
],
[
"Voelker",
"Geoffrey M.",
""
],
[
"Saul",
"Lawrence K.",
""
]
] | TITLE: Topic Modeling of Hierarchical Corpora
ABSTRACT: We study the problem of topic modeling in corpora whose documents are
organized in a multi-level hierarchy. We explore a parametric approach to this
problem, assuming that the number of topics is known or can be estimated by
cross-validation. The models we consider can be viewed as special
(finite-dimensional) instances of hierarchical Dirichlet processes (HDPs). For
these models we show that there exists a simple variational approximation for
probabilistic inference. The approximation relies on a previously unexploited
inequality that handles the conditional dependence between Dirichlet latent
variables in adjacent levels of the model's hierarchy. We compare our approach
to existing implementations of nonparametric HDPs. On several benchmarks we
find that our approach is faster than Gibbs sampling and able to learn more
predictive models than existing variational methods. Finally, we demonstrate
the large-scale viability of our approach on two newly available corpora from
researchers in computer security---one with 350,000 documents and over 6,000
internal subcategories, the other with a five-level deep hierarchy.
| no_new_dataset | 0.708975 |
1412.6597 | Tom Paine | Tom Le Paine, Pooya Khorrami, Wei Han, Thomas S. Huang | An Analysis of Unsupervised Pre-training in Light of Recent Advances | Accepted as a workshop contribution to ICLR 2015 | null | null | null | cs.CV cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Convolutional neural networks perform well on object recognition because of a
number of recent advances: rectified linear units (ReLUs), data augmentation,
dropout, and large labelled datasets. Unsupervised data has been proposed as
another way to improve performance. Unfortunately, unsupervised pre-training is
not used by state-of-the-art methods leading to the following question: Is
unsupervised pre-training still useful given recent advances? If so, when? We
answer this in three parts: we 1) develop an unsupervised method that
incorporates ReLUs and recent unsupervised regularization techniques, 2)
analyze the benefits of unsupervised pre-training compared to data augmentation
and dropout on CIFAR-10 while varying the ratio of unsupervised to supervised
samples, 3) verify our findings on STL-10. We discover unsupervised
pre-training, as expected, helps when the ratio of unsupervised to supervised
samples is high, and surprisingly, hurts when the ratio is low. We also use
unsupervised pre-training with additional color augmentation to achieve near
state-of-the-art performance on STL-10.
| [
{
"version": "v1",
"created": "Sat, 20 Dec 2014 04:20:55 GMT"
},
{
"version": "v2",
"created": "Tue, 27 Jan 2015 22:03:40 GMT"
},
{
"version": "v3",
"created": "Mon, 2 Mar 2015 21:05:34 GMT"
},
{
"version": "v4",
"created": "Fri, 10 Apr 2015 21:26:31 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Paine",
"Tom Le",
""
],
[
"Khorrami",
"Pooya",
""
],
[
"Han",
"Wei",
""
],
[
"Huang",
"Thomas S.",
""
]
] | TITLE: An Analysis of Unsupervised Pre-training in Light of Recent Advances
ABSTRACT: Convolutional neural networks perform well on object recognition because of a
number of recent advances: rectified linear units (ReLUs), data augmentation,
dropout, and large labelled datasets. Unsupervised data has been proposed as
another way to improve performance. Unfortunately, unsupervised pre-training is
not used by state-of-the-art methods leading to the following question: Is
unsupervised pre-training still useful given recent advances? If so, when? We
answer this in three parts: we 1) develop an unsupervised method that
incorporates ReLUs and recent unsupervised regularization techniques, 2)
analyze the benefits of unsupervised pre-training compared to data augmentation
and dropout on CIFAR-10 while varying the ratio of unsupervised to supervised
samples, 3) verify our findings on STL-10. We discover unsupervised
pre-training, as expected, helps when the ratio of unsupervised to supervised
samples is high, and surprisingly, hurts when the ratio is low. We also use
unsupervised pre-training with additional color augmentation to achieve near
state-of-the-art performance on STL-10.
| no_new_dataset | 0.709038 |
1412.6806 | Alexey Dosovitskiy | Jost Tobias Springenberg, Alexey Dosovitskiy, Thomas Brox, Martin
Riedmiller | Striving for Simplicity: The All Convolutional Net | accepted to ICLR-2015 workshop track; no changes other than style | null | null | null | cs.LG cs.CV cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most modern convolutional neural networks (CNNs) used for object recognition
are built using the same principles: Alternating convolution and max-pooling
layers followed by a small number of fully connected layers. We re-evaluate the
state of the art for object recognition from small images with convolutional
networks, questioning the necessity of different components in the pipeline. We
find that max-pooling can simply be replaced by a convolutional layer with
increased stride without loss in accuracy on several image recognition
benchmarks. Following this finding -- and building on other recent work for
finding simple network structures -- we propose a new architecture that
consists solely of convolutional layers and yields competitive or state of the
art performance on several object recognition datasets (CIFAR-10, CIFAR-100,
ImageNet). To analyze the network we introduce a new variant of the
"deconvolution approach" for visualizing features learned by CNNs, which can be
applied to a broader range of network structures than existing approaches.
| [
{
"version": "v1",
"created": "Sun, 21 Dec 2014 16:16:37 GMT"
},
{
"version": "v2",
"created": "Mon, 2 Mar 2015 21:44:06 GMT"
},
{
"version": "v3",
"created": "Mon, 13 Apr 2015 07:58:17 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Springenberg",
"Jost Tobias",
""
],
[
"Dosovitskiy",
"Alexey",
""
],
[
"Brox",
"Thomas",
""
],
[
"Riedmiller",
"Martin",
""
]
] | TITLE: Striving for Simplicity: The All Convolutional Net
ABSTRACT: Most modern convolutional neural networks (CNNs) used for object recognition
are built using the same principles: Alternating convolution and max-pooling
layers followed by a small number of fully connected layers. We re-evaluate the
state of the art for object recognition from small images with convolutional
networks, questioning the necessity of different components in the pipeline. We
find that max-pooling can simply be replaced by a convolutional layer with
increased stride without loss in accuracy on several image recognition
benchmarks. Following this finding -- and building on other recent work for
finding simple network structures -- we propose a new architecture that
consists solely of convolutional layers and yields competitive or state of the
art performance on several object recognition datasets (CIFAR-10, CIFAR-100,
ImageNet). To analyze the network we introduce a new variant of the
"deconvolution approach" for visualizing features learned by CNNs, which can be
applied to a broader range of network structures than existing approaches.
| no_new_dataset | 0.710791 |
1412.8419 | Pedro O. Pinheiro | Remi Lebret and Pedro O. Pinheiro and Ronan Collobert | Simple Image Description Generator via a Linear Phrase-Based Approach | Accepted as a workshop paper at ICLR 2015 | null | null | null | cs.CL cs.CV cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generating a novel textual description of an image is an interesting problem
that connects computer vision and natural language processing. In this paper,
we present a simple model that is able to generate descriptive sentences given
a sample image. This model has a strong focus on the syntax of the
descriptions. We train a purely bilinear model that learns a metric between an
image representation (generated from a previously trained Convolutional Neural
Network) and phrases that are used to described them. The system is then able
to infer phrases from a given image sample. Based on caption syntax statistics,
we propose a simple language model that can produce relevant descriptions for a
given test image using the phrases inferred. Our approach, which is
considerably simpler than state-of-the-art models, achieves comparable results
on the recently release Microsoft COCO dataset.
| [
{
"version": "v1",
"created": "Mon, 29 Dec 2014 18:43:10 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Mar 2015 05:09:13 GMT"
},
{
"version": "v3",
"created": "Sat, 11 Apr 2015 03:53:26 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Lebret",
"Remi",
""
],
[
"Pinheiro",
"Pedro O.",
""
],
[
"Collobert",
"Ronan",
""
]
] | TITLE: Simple Image Description Generator via a Linear Phrase-Based Approach
ABSTRACT: Generating a novel textual description of an image is an interesting problem
that connects computer vision and natural language processing. In this paper,
we present a simple model that is able to generate descriptive sentences given
a sample image. This model has a strong focus on the syntax of the
descriptions. We train a purely bilinear model that learns a metric between an
image representation (generated from a previously trained Convolutional Neural
Network) and phrases that are used to described them. The system is then able
to infer phrases from a given image sample. Based on caption syntax statistics,
we propose a simple language model that can produce relevant descriptions for a
given test image using the phrases inferred. Our approach, which is
considerably simpler than state-of-the-art models, achieves comparable results
on the recently release Microsoft COCO dataset.
| no_new_dataset | 0.708767 |
1501.07180 | Liliang Zhang | Liliang Zhang, Liang Lin, Xian Wu, Shengyong Ding, Lei Zhang | End-to-End Photo-Sketch Generation via Fully Convolutional
Representation Learning | 8 pages, 6 figures. Proceeding in ACM International Conference on
Multimedia Retrieval (ICMR), 2015 | null | 10.1145/2671188.2749321 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sketch-based face recognition is an interesting task in vision and multimedia
research, yet it is quite challenging due to the great difference between face
photos and sketches. In this paper, we propose a novel approach for
photo-sketch generation, aiming to automatically transform face photos into
detail-preserving personal sketches. Unlike the traditional models synthesizing
sketches based on a dictionary of exemplars, we develop a fully convolutional
network to learn the end-to-end photo-sketch mapping. Our approach takes whole
face photos as inputs and directly generates the corresponding sketch images
with efficient inference and learning, in which the architecture are stacked by
only convolutional kernels of very small sizes. To well capture the person
identity during the photo-sketch transformation, we define our optimization
objective in the form of joint generative-discriminative minimization. In
particular, a discriminative regularization term is incorporated into the
photo-sketch generation, enhancing the discriminability of the generated person
sketches against other individuals. Extensive experiments on several standard
benchmarks suggest that our approach outperforms other state-of-the-art methods
in both photo-sketch generation and face sketch verification.
| [
{
"version": "v1",
"created": "Wed, 28 Jan 2015 16:32:53 GMT"
},
{
"version": "v2",
"created": "Sat, 11 Apr 2015 14:28:21 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Zhang",
"Liliang",
""
],
[
"Lin",
"Liang",
""
],
[
"Wu",
"Xian",
""
],
[
"Ding",
"Shengyong",
""
],
[
"Zhang",
"Lei",
""
]
] | TITLE: End-to-End Photo-Sketch Generation via Fully Convolutional
Representation Learning
ABSTRACT: Sketch-based face recognition is an interesting task in vision and multimedia
research, yet it is quite challenging due to the great difference between face
photos and sketches. In this paper, we propose a novel approach for
photo-sketch generation, aiming to automatically transform face photos into
detail-preserving personal sketches. Unlike the traditional models synthesizing
sketches based on a dictionary of exemplars, we develop a fully convolutional
network to learn the end-to-end photo-sketch mapping. Our approach takes whole
face photos as inputs and directly generates the corresponding sketch images
with efficient inference and learning, in which the architecture are stacked by
only convolutional kernels of very small sizes. To well capture the person
identity during the photo-sketch transformation, we define our optimization
objective in the form of joint generative-discriminative minimization. In
particular, a discriminative regularization term is incorporated into the
photo-sketch generation, enhancing the discriminability of the generated person
sketches against other individuals. Extensive experiments on several standard
benchmarks suggest that our approach outperforms other state-of-the-art methods
in both photo-sketch generation and face sketch verification.
| no_new_dataset | 0.71027 |
1502.07645 | Yu-Xiang Wang | Yu-Xiang Wang, Stephen E. Fienberg, Alex Smola | Privacy for Free: Posterior Sampling and Stochastic Gradient Monte Carlo | null | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the problem of Bayesian learning on sensitive datasets and
present two simple but somewhat surprising results that connect Bayesian
learning to "differential privacy:, a cryptographic approach to protect
individual-level privacy while permiting database-level utility. Specifically,
we show that that under standard assumptions, getting one single sample from a
posterior distribution is differentially private "for free". We will see that
estimator is statistically consistent, near optimal and computationally
tractable whenever the Bayesian model of interest is consistent, optimal and
tractable. Similarly but separately, we show that a recent line of works that
use stochastic gradient for Hybrid Monte Carlo (HMC) sampling also preserve
differentially privacy with minor or no modifications of the algorithmic
procedure at all, these observations lead to an "anytime" algorithm for
Bayesian learning under privacy constraint. We demonstrate that it performs
much better than the state-of-the-art differential private methods on synthetic
and real datasets.
| [
{
"version": "v1",
"created": "Thu, 26 Feb 2015 17:38:47 GMT"
},
{
"version": "v2",
"created": "Sun, 12 Apr 2015 02:53:05 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Wang",
"Yu-Xiang",
""
],
[
"Fienberg",
"Stephen E.",
""
],
[
"Smola",
"Alex",
""
]
] | TITLE: Privacy for Free: Posterior Sampling and Stochastic Gradient Monte Carlo
ABSTRACT: We consider the problem of Bayesian learning on sensitive datasets and
present two simple but somewhat surprising results that connect Bayesian
learning to "differential privacy:, a cryptographic approach to protect
individual-level privacy while permiting database-level utility. Specifically,
we show that that under standard assumptions, getting one single sample from a
posterior distribution is differentially private "for free". We will see that
estimator is statistically consistent, near optimal and computationally
tractable whenever the Bayesian model of interest is consistent, optimal and
tractable. Similarly but separately, we show that a recent line of works that
use stochastic gradient for Hybrid Monte Carlo (HMC) sampling also preserve
differentially privacy with minor or no modifications of the algorithmic
procedure at all, these observations lead to an "anytime" algorithm for
Bayesian learning under privacy constraint. We demonstrate that it performs
much better than the state-of-the-art differential private methods on synthetic
and real datasets.
| no_new_dataset | 0.710057 |
1503.02801 | Jiaming Xu | Jiaming Xu, Bo Xu, Guanhua Tian, Jun Zhao, Fangyuan Wang, Hongwei Hao | Short Text Hashing Improved by Integrating Multi-Granularity Topics and
Tags | 12 pages, accepted at CICLing 2015 | null | 10.1007/978-3-319-18111-0_33 | null | cs.IR cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Due to computational and storage efficiencies of compact binary codes,
hashing has been widely used for large-scale similarity search. Unfortunately,
many existing hashing methods based on observed keyword features are not
effective for short texts due to the sparseness and shortness. Recently, some
researchers try to utilize latent topics of certain granularity to preserve
semantic similarity in hash codes beyond keyword matching. However, topics of
certain granularity are not adequate to represent the intrinsic semantic
information. In this paper, we present a novel unified approach for short text
Hashing using Multi-granularity Topics and Tags, dubbed HMTT. In particular, we
propose a selection method to choose the optimal multi-granularity topics
depending on the type of dataset, and design two distinct hashing strategies to
incorporate multi-granularity topics. We also propose a simple and effective
method to exploit tags to enhance the similarity of related texts. We carry out
extensive experiments on one short text dataset as well as on one normal text
dataset. The results demonstrate that our approach is effective and
significantly outperforms baselines on several evaluation metrics.
| [
{
"version": "v1",
"created": "Tue, 10 Mar 2015 07:51:59 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Xu",
"Jiaming",
""
],
[
"Xu",
"Bo",
""
],
[
"Tian",
"Guanhua",
""
],
[
"Zhao",
"Jun",
""
],
[
"Wang",
"Fangyuan",
""
],
[
"Hao",
"Hongwei",
""
]
] | TITLE: Short Text Hashing Improved by Integrating Multi-Granularity Topics and
Tags
ABSTRACT: Due to computational and storage efficiencies of compact binary codes,
hashing has been widely used for large-scale similarity search. Unfortunately,
many existing hashing methods based on observed keyword features are not
effective for short texts due to the sparseness and shortness. Recently, some
researchers try to utilize latent topics of certain granularity to preserve
semantic similarity in hash codes beyond keyword matching. However, topics of
certain granularity are not adequate to represent the intrinsic semantic
information. In this paper, we present a novel unified approach for short text
Hashing using Multi-granularity Topics and Tags, dubbed HMTT. In particular, we
propose a selection method to choose the optimal multi-granularity topics
depending on the type of dataset, and design two distinct hashing strategies to
incorporate multi-granularity topics. We also propose a simple and effective
method to exploit tags to enhance the similarity of related texts. We carry out
extensive experiments on one short text dataset as well as on one normal text
dataset. The results demonstrate that our approach is effective and
significantly outperforms baselines on several evaluation metrics.
| no_new_dataset | 0.70883 |
1503.06813 | Tarek El-Gaaly | Haopeng Zhang, Tarek El-Gaaly, Ahmed Elgammal, Zhiguo Jiang | Factorization of View-Object Manifolds for Joint Object Recognition and
Pose Estimation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Due to large variations in shape, appearance, and viewing conditions, object
recognition is a key precursory challenge in the fields of object manipulation
and robotic/AI visual reasoning in general. Recognizing object categories,
particular instances of objects and viewpoints/poses of objects are three
critical subproblems robots must solve in order to accurately grasp/manipulate
objects and reason about their environments. Multi-view images of the same
object lie on intrinsic low-dimensional manifolds in descriptor spaces (e.g.
visual/depth descriptor spaces). These object manifolds share the same topology
despite being geometrically different. Each object manifold can be represented
as a deformed version of a unified manifold. The object manifolds can thus be
parameterized by its homeomorphic mapping/reconstruction from the unified
manifold. In this work, we develop a novel framework to jointly solve the three
challenging recognition sub-problems, by explicitly modeling the deformations
of object manifolds and factorizing it in a view-invariant space for
recognition. We perform extensive experiments on several challenging datasets
and achieve state-of-the-art results.
| [
{
"version": "v1",
"created": "Mon, 23 Mar 2015 20:05:36 GMT"
},
{
"version": "v2",
"created": "Mon, 13 Apr 2015 02:59:41 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Zhang",
"Haopeng",
""
],
[
"El-Gaaly",
"Tarek",
""
],
[
"Elgammal",
"Ahmed",
""
],
[
"Jiang",
"Zhiguo",
""
]
] | TITLE: Factorization of View-Object Manifolds for Joint Object Recognition and
Pose Estimation
ABSTRACT: Due to large variations in shape, appearance, and viewing conditions, object
recognition is a key precursory challenge in the fields of object manipulation
and robotic/AI visual reasoning in general. Recognizing object categories,
particular instances of objects and viewpoints/poses of objects are three
critical subproblems robots must solve in order to accurately grasp/manipulate
objects and reason about their environments. Multi-view images of the same
object lie on intrinsic low-dimensional manifolds in descriptor spaces (e.g.
visual/depth descriptor spaces). These object manifolds share the same topology
despite being geometrically different. Each object manifold can be represented
as a deformed version of a unified manifold. The object manifolds can thus be
parameterized by its homeomorphic mapping/reconstruction from the unified
manifold. In this work, we develop a novel framework to jointly solve the three
challenging recognition sub-problems, by explicitly modeling the deformations
of object manifolds and factorizing it in a view-invariant space for
recognition. We perform extensive experiments on several challenging datasets
and achieve state-of-the-art results.
| no_new_dataset | 0.710465 |
1503.08909 | George Toderici | Joe Yue-Hei Ng and Matthew Hausknecht and Sudheendra Vijayanarasimhan
and Oriol Vinyals and Rajat Monga and George Toderici | Beyond Short Snippets: Deep Networks for Video Classification | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Convolutional neural networks (CNNs) have been extensively applied for image
recognition problems giving state-of-the-art results on recognition, detection,
segmentation and retrieval. In this work we propose and evaluate several deep
neural network architectures to combine image information across a video over
longer time periods than previously attempted. We propose two methods capable
of handling full length videos. The first method explores various convolutional
temporal feature pooling architectures, examining the various design choices
which need to be made when adapting a CNN for this task. The second proposed
method explicitly models the video as an ordered sequence of frames. For this
purpose we employ a recurrent neural network that uses Long Short-Term Memory
(LSTM) cells which are connected to the output of the underlying CNN. Our best
networks exhibit significant performance improvements over previously published
results on the Sports 1 million dataset (73.1% vs. 60.9%) and the UCF-101
datasets with (88.6% vs. 88.0%) and without additional optical flow information
(82.6% vs. 72.8%).
| [
{
"version": "v1",
"created": "Tue, 31 Mar 2015 04:34:12 GMT"
},
{
"version": "v2",
"created": "Mon, 13 Apr 2015 19:44:25 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Ng",
"Joe Yue-Hei",
""
],
[
"Hausknecht",
"Matthew",
""
],
[
"Vijayanarasimhan",
"Sudheendra",
""
],
[
"Vinyals",
"Oriol",
""
],
[
"Monga",
"Rajat",
""
],
[
"Toderici",
"George",
""
]
] | TITLE: Beyond Short Snippets: Deep Networks for Video Classification
ABSTRACT: Convolutional neural networks (CNNs) have been extensively applied for image
recognition problems giving state-of-the-art results on recognition, detection,
segmentation and retrieval. In this work we propose and evaluate several deep
neural network architectures to combine image information across a video over
longer time periods than previously attempted. We propose two methods capable
of handling full length videos. The first method explores various convolutional
temporal feature pooling architectures, examining the various design choices
which need to be made when adapting a CNN for this task. The second proposed
method explicitly models the video as an ordered sequence of frames. For this
purpose we employ a recurrent neural network that uses Long Short-Term Memory
(LSTM) cells which are connected to the output of the underlying CNN. Our best
networks exhibit significant performance improvements over previously published
results on the Sports 1 million dataset (73.1% vs. 60.9%) and the UCF-101
datasets with (88.6% vs. 88.0%) and without additional optical flow information
(82.6% vs. 72.8%).
| no_new_dataset | 0.712432 |
1504.00591 | Jonathan Beard | Jonathan C. Beard and Roger D. Chamberlain | Run Time Approximation of Non-blocking Service Rates for Streaming
Systems | technical report | null | null | null | cs.PF | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Stream processing is a compute paradigm that promises safe and efficient
parallelism. Modern big-data problems are often well suited for stream
processing's throughput-oriented nature. Realization of efficient stream
processing requires monitoring and optimization of multiple communications
links. Most techniques to optimize these links use queueing network models or
network flow models, which require some idea of the actual execution rate of
each independent compute kernel within the system. What we want to know is how
fast can each kernel process data independent of other communicating kernels.
This is known as the "service rate" of the kernel within the queueing
literature. Current approaches to divining service rates are static. Modern
workloads, however, are often dynamic. Shared cloud systems also present
applications with highly dynamic execution environments (multiple users,
hardware migration, etc.). It is therefore desirable to continuously re-tune an
application during run time (online) in response to changing conditions. Our
approach enables online service rate monitoring under most conditions,
obviating the need for reliance on steady state predictions for what are
probably non-steady state phenomena. First, some of the difficulties associated
with online service rate determination are examined. Second, the algorithm to
approximate the online non-blocking service rate is described. Lastly, the
algorithm is implemented within the open source RaftLib framework for
validation using a simple microbenchmark as well as two full streaming
applications.
| [
{
"version": "v1",
"created": "Thu, 2 Apr 2015 15:47:19 GMT"
},
{
"version": "v2",
"created": "Mon, 13 Apr 2015 00:08:31 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Beard",
"Jonathan C.",
""
],
[
"Chamberlain",
"Roger D.",
""
]
] | TITLE: Run Time Approximation of Non-blocking Service Rates for Streaming
Systems
ABSTRACT: Stream processing is a compute paradigm that promises safe and efficient
parallelism. Modern big-data problems are often well suited for stream
processing's throughput-oriented nature. Realization of efficient stream
processing requires monitoring and optimization of multiple communications
links. Most techniques to optimize these links use queueing network models or
network flow models, which require some idea of the actual execution rate of
each independent compute kernel within the system. What we want to know is how
fast can each kernel process data independent of other communicating kernels.
This is known as the "service rate" of the kernel within the queueing
literature. Current approaches to divining service rates are static. Modern
workloads, however, are often dynamic. Shared cloud systems also present
applications with highly dynamic execution environments (multiple users,
hardware migration, etc.). It is therefore desirable to continuously re-tune an
application during run time (online) in response to changing conditions. Our
approach enables online service rate monitoring under most conditions,
obviating the need for reliance on steady state predictions for what are
probably non-steady state phenomena. First, some of the difficulties associated
with online service rate determination are examined. Second, the algorithm to
approximate the online non-blocking service rate is described. Lastly, the
algorithm is implemented within the open source RaftLib framework for
validation using a simple microbenchmark as well as two full streaming
applications.
| no_new_dataset | 0.707019 |
1504.02886 | Minghua Xia | Minghua Xia and Sonia A\"issa | Fundamental Relations Between Reactive and Proactive Relay-Selection
Strategies | 4 pages, to appear in IEEE Communications Letters | null | null | null | cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Two major relay-selection strategies widely applied in cooperative
decode-and-forward (DF) relaying networks, namely, reactive relay selection
(RRS) and proactive relay selection (PRS), are generally looked upon as
independent and studied separately. In this paper, RRS and PRS are proven to be
equivalent with respect to the end-to-end outage probability from the first
principle, i.e. their respective relay-selection criteria. On the other hand,
RRS is shown to be superior to PRS with respect to the end-to-end symbol error
rate. Afterwards, a case study of a general DF relaying system, subject to
co-channel interferences and additive white Gaussian noise at both the relaying
nodes and the destination, is performed to explicitly illustrate the
aforementioned outage equivalence. These fundamental relations provide
intuitive yet insightful performance benchmarks for comparing various
applications of these two relay-selection strategies.
| [
{
"version": "v1",
"created": "Sat, 11 Apr 2015 15:14:48 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Xia",
"Minghua",
""
],
[
"Aïssa",
"Sonia",
""
]
] | TITLE: Fundamental Relations Between Reactive and Proactive Relay-Selection
Strategies
ABSTRACT: Two major relay-selection strategies widely applied in cooperative
decode-and-forward (DF) relaying networks, namely, reactive relay selection
(RRS) and proactive relay selection (PRS), are generally looked upon as
independent and studied separately. In this paper, RRS and PRS are proven to be
equivalent with respect to the end-to-end outage probability from the first
principle, i.e. their respective relay-selection criteria. On the other hand,
RRS is shown to be superior to PRS with respect to the end-to-end symbol error
rate. Afterwards, a case study of a general DF relaying system, subject to
co-channel interferences and additive white Gaussian noise at both the relaying
nodes and the destination, is performed to explicitly illustrate the
aforementioned outage equivalence. These fundamental relations provide
intuitive yet insightful performance benchmarks for comparing various
applications of these two relay-selection strategies.
| no_new_dataset | 0.711825 |
1504.02902 | Alexander Kalmanovich | Alexander Kalmanovich and Gal Chechik | Gradual Training Method for Denoising Auto Encoders | arXiv admin note: substantial text overlap with arXiv:1412.6257 | null | null | null | cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Stacked denoising auto encoders (DAEs) are well known to learn useful deep
representations, which can be used to improve supervised training by
initializing a deep network. We investigate a training scheme of a deep DAE,
where DAE layers are gradually added and keep adapting as additional layers are
added. We show that in the regime of mid-sized datasets, this gradual training
provides a small but consistent improvement over stacked training in both
reconstruction quality and classification error over stacked training on MNIST
and CIFAR datasets.
| [
{
"version": "v1",
"created": "Sat, 11 Apr 2015 17:51:41 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Kalmanovich",
"Alexander",
""
],
[
"Chechik",
"Gal",
""
]
] | TITLE: Gradual Training Method for Denoising Auto Encoders
ABSTRACT: Stacked denoising auto encoders (DAEs) are well known to learn useful deep
representations, which can be used to improve supervised training by
initializing a deep network. We investigate a training scheme of a deep DAE,
where DAE layers are gradually added and keep adapting as additional layers are
added. We show that in the regime of mid-sized datasets, this gradual training
provides a small but consistent improvement over stacked training in both
reconstruction quality and classification error over stacked training on MNIST
and CIFAR datasets.
| no_new_dataset | 0.710785 |
1504.02975 | F. Ozgur Catak | Ferhat \"Ozg\"ur \c{C}atak | Classification with Extreme Learning Machine and Ensemble Algorithms
Over Randomly Partitioned Data | In Turkish, SIU | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this age of Big Data, machine learning based data mining methods are
extensively used to inspect large scale data sets. Deriving applicable
predictive modeling from these type of data sets is a challenging obstacle
because of their high complexity. Opportunity with high data availability
levels, automated classification of data sets has become a critical and
complicated function. In this paper, the power of applying MapReduce based
Distributed AdaBoosting of Extreme Learning Machine (ELM) are explored to build
reliable predictive bag of classification models. Thus, (i) dataset ensembles
are build; (ii) ELM algorithm is used to build weak classification models; and
(iii) build a strong classification model from a set of weak classification
models. This training model is applied to the publicly available knowledge
discovery and data mining datasets.
| [
{
"version": "v1",
"created": "Sun, 12 Apr 2015 14:03:25 GMT"
}
] | 2015-04-14T00:00:00 | [
[
"Çatak",
"Ferhat Özgür",
""
]
] | TITLE: Classification with Extreme Learning Machine and Ensemble Algorithms
Over Randomly Partitioned Data
ABSTRACT: In this age of Big Data, machine learning based data mining methods are
extensively used to inspect large scale data sets. Deriving applicable
predictive modeling from these type of data sets is a challenging obstacle
because of their high complexity. Opportunity with high data availability
levels, automated classification of data sets has become a critical and
complicated function. In this paper, the power of applying MapReduce based
Distributed AdaBoosting of Extreme Learning Machine (ELM) are explored to build
reliable predictive bag of classification models. Thus, (i) dataset ensembles
are build; (ii) ELM algorithm is used to build weak classification models; and
(iii) build a strong classification model from a set of weak classification
models. This training model is applied to the publicly available knowledge
discovery and data mining datasets.
| no_new_dataset | 0.709044 |
1301.1275 | Luca Allodi | Luca Allodi, Fabio Massacci | My Software has a Vulnerability, should I worry? | 12 pages, 4 figures | ACM TISSEC Vol 17 Issue 1, 2014 | 10.1145/2630069 | null | cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | (U.S) Rule-based policies to mitigate software risk suggest to use the CVSS
score to measure the individual vulnerability risk and act accordingly: an HIGH
CVSS score according to the NVD (National (U.S.) Vulnerability Database) is
therefore translated into a "Yes". A key issue is whether such rule is
economically sensible, in particular if reported vulnerabilities have been
actually exploited in the wild, and whether the risk score do actually match
the risk of actual exploitation.
We compare the NVD dataset with two additional datasets, the EDB for the
white market of vulnerabilities (such as those present in Metasploit), and the
EKITS for the exploits traded in the black market. We benchmark them against
Symantec's threat explorer dataset (SYM) of actual exploit in the wild. We
analyze the whole spectrum of CVSS submetrics and use these characteristics to
perform a case-controlled analysis of CVSS scores (similar to those used to
link lung cancer and smoking) to test its reliability as a risk factor for
actual exploitation.
We conclude that (a) fixing just because a high CVSS score in NVD only yields
negligible risk reduction, (b) the additional existence of proof of concepts
exploits (e.g. in EDB) may yield some additional but not large risk reduction,
(c) fixing in response to presence in black markets yields the equivalent risk
reduction of wearing safety belt in cars (you might also die but still..). On
the negative side, our study shows that as industry we miss a metric with high
specificity (ruling out vulns for which we shouldn't worry).
In order to address the feedback from BlackHat 2013's audience, the final
revision (V3) provides additional data in Appendix A detailing how the control
variables in the study affect the results.
| [
{
"version": "v1",
"created": "Mon, 7 Jan 2013 17:32:36 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Aug 2013 21:09:41 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Sep 2013 14:05:27 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Allodi",
"Luca",
""
],
[
"Massacci",
"Fabio",
""
]
] | TITLE: My Software has a Vulnerability, should I worry?
ABSTRACT: (U.S) Rule-based policies to mitigate software risk suggest to use the CVSS
score to measure the individual vulnerability risk and act accordingly: an HIGH
CVSS score according to the NVD (National (U.S.) Vulnerability Database) is
therefore translated into a "Yes". A key issue is whether such rule is
economically sensible, in particular if reported vulnerabilities have been
actually exploited in the wild, and whether the risk score do actually match
the risk of actual exploitation.
We compare the NVD dataset with two additional datasets, the EDB for the
white market of vulnerabilities (such as those present in Metasploit), and the
EKITS for the exploits traded in the black market. We benchmark them against
Symantec's threat explorer dataset (SYM) of actual exploit in the wild. We
analyze the whole spectrum of CVSS submetrics and use these characteristics to
perform a case-controlled analysis of CVSS scores (similar to those used to
link lung cancer and smoking) to test its reliability as a risk factor for
actual exploitation.
We conclude that (a) fixing just because a high CVSS score in NVD only yields
negligible risk reduction, (b) the additional existence of proof of concepts
exploits (e.g. in EDB) may yield some additional but not large risk reduction,
(c) fixing in response to presence in black markets yields the equivalent risk
reduction of wearing safety belt in cars (you might also die but still..). On
the negative side, our study shows that as industry we miss a metric with high
specificity (ruling out vulns for which we shouldn't worry).
In order to address the feedback from BlackHat 2013's audience, the final
revision (V3) provides additional data in Appendix A detailing how the control
variables in the study affect the results.
| no_new_dataset | 0.691224 |
1406.2080 | Sainbayar Sukhbaatar | Sainbayar Sukhbaatar, Joan Bruna, Manohar Paluri, Lubomir Bourdev and
Rob Fergus | Training Convolutional Networks with Noisy Labels | Accepted as a workshop contribution at ICLR 2015 | null | null | null | cs.CV cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The availability of large labeled datasets has allowed Convolutional Network
models to achieve impressive recognition results. However, in many settings
manual annotation of the data is impractical; instead our data has noisy
labels, i.e. there is some freely available label for each image which may or
may not be accurate. In this paper, we explore the performance of
discriminatively-trained Convnets when trained on such noisy data. We introduce
an extra noise layer into the network which adapts the network outputs to match
the noisy label distribution. The parameters of this noise layer can be
estimated as part of the training process and involve simple modifications to
current training infrastructures for deep networks. We demonstrate the
approaches on several datasets, including large scale experiments on the
ImageNet classification benchmark.
| [
{
"version": "v1",
"created": "Mon, 9 Jun 2014 05:45:12 GMT"
},
{
"version": "v2",
"created": "Sat, 20 Dec 2014 21:10:03 GMT"
},
{
"version": "v3",
"created": "Tue, 3 Mar 2015 21:13:47 GMT"
},
{
"version": "v4",
"created": "Fri, 10 Apr 2015 16:44:00 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Sukhbaatar",
"Sainbayar",
""
],
[
"Bruna",
"Joan",
""
],
[
"Paluri",
"Manohar",
""
],
[
"Bourdev",
"Lubomir",
""
],
[
"Fergus",
"Rob",
""
]
] | TITLE: Training Convolutional Networks with Noisy Labels
ABSTRACT: The availability of large labeled datasets has allowed Convolutional Network
models to achieve impressive recognition results. However, in many settings
manual annotation of the data is impractical; instead our data has noisy
labels, i.e. there is some freely available label for each image which may or
may not be accurate. In this paper, we explore the performance of
discriminatively-trained Convnets when trained on such noisy data. We introduce
an extra noise layer into the network which adapts the network outputs to match
the noisy label distribution. The parameters of this noise layer can be
estimated as part of the training process and involve simple modifications to
current training infrastructures for deep networks. We demonstrate the
approaches on several datasets, including large scale experiments on the
ImageNet classification benchmark.
| no_new_dataset | 0.710446 |
1409.1556 | Karen Simonyan | Karen Simonyan, Andrew Zisserman | Very Deep Convolutional Networks for Large-Scale Image Recognition | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work we investigate the effect of the convolutional network depth on
its accuracy in the large-scale image recognition setting. Our main
contribution is a thorough evaluation of networks of increasing depth using an
architecture with very small (3x3) convolution filters, which shows that a
significant improvement on the prior-art configurations can be achieved by
pushing the depth to 16-19 weight layers. These findings were the basis of our
ImageNet Challenge 2014 submission, where our team secured the first and the
second places in the localisation and classification tracks respectively. We
also show that our representations generalise well to other datasets, where
they achieve state-of-the-art results. We have made our two best-performing
ConvNet models publicly available to facilitate further research on the use of
deep visual representations in computer vision.
| [
{
"version": "v1",
"created": "Thu, 4 Sep 2014 19:48:04 GMT"
},
{
"version": "v2",
"created": "Mon, 15 Sep 2014 19:58:29 GMT"
},
{
"version": "v3",
"created": "Tue, 18 Nov 2014 20:43:11 GMT"
},
{
"version": "v4",
"created": "Fri, 19 Dec 2014 20:01:21 GMT"
},
{
"version": "v5",
"created": "Tue, 23 Dec 2014 20:05:00 GMT"
},
{
"version": "v6",
"created": "Fri, 10 Apr 2015 16:25:04 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Simonyan",
"Karen",
""
],
[
"Zisserman",
"Andrew",
""
]
] | TITLE: Very Deep Convolutional Networks for Large-Scale Image Recognition
ABSTRACT: In this work we investigate the effect of the convolutional network depth on
its accuracy in the large-scale image recognition setting. Our main
contribution is a thorough evaluation of networks of increasing depth using an
architecture with very small (3x3) convolution filters, which shows that a
significant improvement on the prior-art configurations can be achieved by
pushing the depth to 16-19 weight layers. These findings were the basis of our
ImageNet Challenge 2014 submission, where our team secured the first and the
second places in the localisation and classification tracks respectively. We
also show that our representations generalise well to other datasets, where
they achieve state-of-the-art results. We have made our two best-performing
ConvNet models publicly available to facilitate further research on the use of
deep visual representations in computer vision.
| no_new_dataset | 0.711625 |
1410.1165 | Rupesh Kumar Srivastava | Rupesh Kumar Srivastava, Jonathan Masci, Faustino Gomez, J\"urgen
Schmidhuber | Understanding Locally Competitive Networks | 9 pages + 2 supplementary, Accepted to ICLR 2015 Conference track | null | null | null | cs.NE cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently proposed neural network activation functions such as rectified
linear, maxout, and local winner-take-all have allowed for faster and more
effective training of deep neural architectures on large and complex datasets.
The common trait among these functions is that they implement local competition
between small groups of computational units within a layer, so that only part
of the network is activated for any given input pattern. In this paper, we
attempt to visualize and understand this self-modularization, and suggest a
unified explanation for the beneficial properties of such networks. We also
show how our insights can be directly useful for efficiently performing
retrieval over large datasets using neural networks.
| [
{
"version": "v1",
"created": "Sun, 5 Oct 2014 14:46:47 GMT"
},
{
"version": "v2",
"created": "Mon, 22 Dec 2014 20:07:17 GMT"
},
{
"version": "v3",
"created": "Thu, 9 Apr 2015 01:22:49 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Srivastava",
"Rupesh Kumar",
""
],
[
"Masci",
"Jonathan",
""
],
[
"Gomez",
"Faustino",
""
],
[
"Schmidhuber",
"Jürgen",
""
]
] | TITLE: Understanding Locally Competitive Networks
ABSTRACT: Recently proposed neural network activation functions such as rectified
linear, maxout, and local winner-take-all have allowed for faster and more
effective training of deep neural architectures on large and complex datasets.
The common trait among these functions is that they implement local competition
between small groups of computational units within a layer, so that only part
of the network is activated for any given input pattern. In this paper, we
attempt to visualize and understand this self-modularization, and suggest a
unified explanation for the beneficial properties of such networks. We also
show how our insights can be directly useful for efficiently performing
retrieval over large datasets using neural networks.
| no_new_dataset | 0.710176 |
1410.8516 | Laurent Dinh | Laurent Dinh, David Krueger and Yoshua Bengio | NICE: Non-linear Independent Components Estimation | 11 pages and 2 pages Appendix, workshop paper at ICLR 2015 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a deep learning framework for modeling complex high-dimensional
densities called Non-linear Independent Component Estimation (NICE). It is
based on the idea that a good representation is one in which the data has a
distribution that is easy to model. For this purpose, a non-linear
deterministic transformation of the data is learned that maps it to a latent
space so as to make the transformed data conform to a factorized distribution,
i.e., resulting in independent latent variables. We parametrize this
transformation so that computing the Jacobian determinant and inverse transform
is trivial, yet we maintain the ability to learn complex non-linear
transformations, via a composition of simple building blocks, each based on a
deep neural network. The training criterion is simply the exact log-likelihood,
which is tractable. Unbiased ancestral sampling is also easy. We show that this
approach yields good generative models on four image datasets and can be used
for inpainting.
| [
{
"version": "v1",
"created": "Thu, 30 Oct 2014 19:44:20 GMT"
},
{
"version": "v2",
"created": "Fri, 19 Dec 2014 22:40:18 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jan 2015 18:10:44 GMT"
},
{
"version": "v4",
"created": "Mon, 9 Mar 2015 18:06:58 GMT"
},
{
"version": "v5",
"created": "Thu, 12 Mar 2015 06:25:20 GMT"
},
{
"version": "v6",
"created": "Fri, 10 Apr 2015 12:27:56 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Dinh",
"Laurent",
""
],
[
"Krueger",
"David",
""
],
[
"Bengio",
"Yoshua",
""
]
] | TITLE: NICE: Non-linear Independent Components Estimation
ABSTRACT: We propose a deep learning framework for modeling complex high-dimensional
densities called Non-linear Independent Component Estimation (NICE). It is
based on the idea that a good representation is one in which the data has a
distribution that is easy to model. For this purpose, a non-linear
deterministic transformation of the data is learned that maps it to a latent
space so as to make the transformed data conform to a factorized distribution,
i.e., resulting in independent latent variables. We parametrize this
transformation so that computing the Jacobian determinant and inverse transform
is trivial, yet we maintain the ability to learn complex non-linear
transformations, via a composition of simple building blocks, each based on a
deep neural network. The training criterion is simply the exact log-likelihood,
which is tractable. Unbiased ancestral sampling is also easy. We show that this
approach yields good generative models on four image datasets and can be used
for inpainting.
| no_new_dataset | 0.70878 |
1412.5903 | Max Jaderberg | Max Jaderberg, Karen Simonyan, Andrea Vedaldi, Andrew Zisserman | Deep Structured Output Learning for Unconstrained Text Recognition | arXiv admin note: text overlap with arXiv:1406.2227 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We develop a representation suitable for the unconstrained recognition of
words in natural images: the general case of no fixed lexicon and unknown
length.
To this end we propose a convolutional neural network (CNN) based
architecture which incorporates a Conditional Random Field (CRF) graphical
model, taking the whole word image as a single input. The unaries of the CRF
are provided by a CNN that predicts characters at each position of the output,
while higher order terms are provided by another CNN that detects the presence
of N-grams. We show that this entire model (CRF, character predictor, N-gram
predictor) can be jointly optimised by back-propagating the structured output
loss, essentially requiring the system to perform multi-task learning, and
training uses purely synthetically generated data. The resulting model is a
more accurate system on standard real-world text recognition benchmarks than
character prediction alone, setting a benchmark for systems that have not been
trained on a particular lexicon. In addition, our model achieves
state-of-the-art accuracy in lexicon-constrained scenarios, without being
specifically modelled for constrained recognition. To test the generalisation
of our model, we also perform experiments with random alpha-numeric strings to
evaluate the method when no visual language model is applicable.
| [
{
"version": "v1",
"created": "Thu, 18 Dec 2014 15:49:46 GMT"
},
{
"version": "v2",
"created": "Fri, 19 Dec 2014 17:37:37 GMT"
},
{
"version": "v3",
"created": "Mon, 22 Dec 2014 19:56:48 GMT"
},
{
"version": "v4",
"created": "Tue, 23 Dec 2014 13:17:59 GMT"
},
{
"version": "v5",
"created": "Fri, 10 Apr 2015 15:36:01 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Jaderberg",
"Max",
""
],
[
"Simonyan",
"Karen",
""
],
[
"Vedaldi",
"Andrea",
""
],
[
"Zisserman",
"Andrew",
""
]
] | TITLE: Deep Structured Output Learning for Unconstrained Text Recognition
ABSTRACT: We develop a representation suitable for the unconstrained recognition of
words in natural images: the general case of no fixed lexicon and unknown
length.
To this end we propose a convolutional neural network (CNN) based
architecture which incorporates a Conditional Random Field (CRF) graphical
model, taking the whole word image as a single input. The unaries of the CRF
are provided by a CNN that predicts characters at each position of the output,
while higher order terms are provided by another CNN that detects the presence
of N-grams. We show that this entire model (CRF, character predictor, N-gram
predictor) can be jointly optimised by back-propagating the structured output
loss, essentially requiring the system to perform multi-task learning, and
training uses purely synthetically generated data. The resulting model is a
more accurate system on standard real-world text recognition benchmarks than
character prediction alone, setting a benchmark for systems that have not been
trained on a particular lexicon. In addition, our model achieves
state-of-the-art accuracy in lexicon-constrained scenarios, without being
specifically modelled for constrained recognition. To test the generalisation
of our model, we also perform experiments with random alpha-numeric strings to
evaluate the method when no visual language model is applicable.
| no_new_dataset | 0.710666 |
1503.05032 | Weifeng Liu | Weifeng Liu, Brian Vinter | CSR5: An Efficient Storage Format for Cross-Platform Sparse
Matrix-Vector Multiplication | 12 pages, 10 figures, In Proceedings of the 29th ACM International
Conference on Supercomputing (ICS '15) | null | null | null | cs.MS cs.DC math.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sparse matrix-vector multiplication (SpMV) is a fundamental building block
for numerous applications. In this paper, we propose CSR5 (Compressed Sparse
Row 5), a new storage format, which offers high-throughput SpMV on various
platforms including CPUs, GPUs and Xeon Phi. First, the CSR5 format is
insensitive to the sparsity structure of the input matrix. Thus the single
format can support an SpMV algorithm that is efficient both for regular
matrices and for irregular matrices. Furthermore, we show that the overhead of
the format conversion from the CSR to the CSR5 can be as low as the cost of a
few SpMV operations. We compare the CSR5-based SpMV algorithm with 11
state-of-the-art formats and algorithms on four mainstream processors using 14
regular and 10 irregular matrices as a benchmark suite. For the 14 regular
matrices in the suite, we achieve comparable or better performance over the
previous work. For the 10 irregular matrices, the CSR5 obtains average
performance improvement of 17.6\%, 28.5\%, 173.0\% and 293.3\% (up to 213.3\%,
153.6\%, 405.1\% and 943.3\%) over the best existing work on dual-socket Intel
CPUs, an nVidia GPU, an AMD GPU and an Intel Xeon Phi, respectively. For
real-world applications such as a solver with only tens of iterations, the CSR5
format can be more practical because of its low-overhead for format conversion.
The source code of this work is downloadable at
https://github.com/bhSPARSE/Benchmark_SpMV_using_CSR5
| [
{
"version": "v1",
"created": "Tue, 17 Mar 2015 13:18:49 GMT"
},
{
"version": "v2",
"created": "Thu, 9 Apr 2015 21:28:03 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Liu",
"Weifeng",
""
],
[
"Vinter",
"Brian",
""
]
] | TITLE: CSR5: An Efficient Storage Format for Cross-Platform Sparse
Matrix-Vector Multiplication
ABSTRACT: Sparse matrix-vector multiplication (SpMV) is a fundamental building block
for numerous applications. In this paper, we propose CSR5 (Compressed Sparse
Row 5), a new storage format, which offers high-throughput SpMV on various
platforms including CPUs, GPUs and Xeon Phi. First, the CSR5 format is
insensitive to the sparsity structure of the input matrix. Thus the single
format can support an SpMV algorithm that is efficient both for regular
matrices and for irregular matrices. Furthermore, we show that the overhead of
the format conversion from the CSR to the CSR5 can be as low as the cost of a
few SpMV operations. We compare the CSR5-based SpMV algorithm with 11
state-of-the-art formats and algorithms on four mainstream processors using 14
regular and 10 irregular matrices as a benchmark suite. For the 14 regular
matrices in the suite, we achieve comparable or better performance over the
previous work. For the 10 irregular matrices, the CSR5 obtains average
performance improvement of 17.6\%, 28.5\%, 173.0\% and 293.3\% (up to 213.3\%,
153.6\%, 405.1\% and 943.3\%) over the best existing work on dual-socket Intel
CPUs, an nVidia GPU, an AMD GPU and an Intel Xeon Phi, respectively. For
real-world applications such as a solver with only tens of iterations, the CSR5
format can be more practical because of its low-overhead for format conversion.
The source code of this work is downloadable at
https://github.com/bhSPARSE/Benchmark_SpMV_using_CSR5
| no_new_dataset | 0.707847 |
1503.08663 | Guanbin Li | Guanbin Li and Yizhou Yu | Visual Saliency Based on Multiscale Deep Features | To appear in CVPR 2015 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual saliency is a fundamental problem in both cognitive and computational
sciences, including computer vision. In this CVPR 2015 paper, we discover that
a high-quality visual saliency model can be trained with multiscale features
extracted using a popular deep learning architecture, convolutional neural
networks (CNNs), which have had many successes in visual recognition tasks. For
learning such saliency models, we introduce a neural network architecture,
which has fully connected layers on top of CNNs responsible for extracting
features at three different scales. We then propose a refinement method to
enhance the spatial coherence of our saliency results. Finally, aggregating
multiple saliency maps computed for different levels of image segmentation can
further boost the performance, yielding saliency maps better than those
generated from a single segmentation. To promote further research and
evaluation of visual saliency models, we also construct a new large database of
4447 challenging images and their pixelwise saliency annotation. Experimental
results demonstrate that our proposed method is capable of achieving
state-of-the-art performance on all public benchmarks, improving the F-Measure
by 5.0% and 13.2% respectively on the MSRA-B dataset and our new dataset
(HKU-IS), and lowering the mean absolute error by 5.7% and 35.1% respectively
on these two datasets.
| [
{
"version": "v1",
"created": "Mon, 30 Mar 2015 13:21:09 GMT"
},
{
"version": "v2",
"created": "Tue, 31 Mar 2015 05:21:02 GMT"
},
{
"version": "v3",
"created": "Fri, 10 Apr 2015 06:40:46 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Li",
"Guanbin",
""
],
[
"Yu",
"Yizhou",
""
]
] | TITLE: Visual Saliency Based on Multiscale Deep Features
ABSTRACT: Visual saliency is a fundamental problem in both cognitive and computational
sciences, including computer vision. In this CVPR 2015 paper, we discover that
a high-quality visual saliency model can be trained with multiscale features
extracted using a popular deep learning architecture, convolutional neural
networks (CNNs), which have had many successes in visual recognition tasks. For
learning such saliency models, we introduce a neural network architecture,
which has fully connected layers on top of CNNs responsible for extracting
features at three different scales. We then propose a refinement method to
enhance the spatial coherence of our saliency results. Finally, aggregating
multiple saliency maps computed for different levels of image segmentation can
further boost the performance, yielding saliency maps better than those
generated from a single segmentation. To promote further research and
evaluation of visual saliency models, we also construct a new large database of
4447 challenging images and their pixelwise saliency annotation. Experimental
results demonstrate that our proposed method is capable of achieving
state-of-the-art performance on all public benchmarks, improving the F-Measure
by 5.0% and 13.2% respectively on the MSRA-B dataset and our new dataset
(HKU-IS), and lowering the mean absolute error by 5.7% and 35.1% respectively
on these two datasets.
| new_dataset | 0.715815 |
1504.02564 | Ragesh Jaiswal | Anup Bhattacharya, Ragesh Jaiswal, Amit Kumar | Faster Algorithms for the Constrained k-means Problem | null | null | null | null | cs.DS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The classical center based clustering problems such as
$k$-means/median/center assume that the optimal clusters satisfy the locality
property that the points in the same cluster are close to each other. A number
of clustering problems arise in machine learning where the optimal clusters do
not follow such a locality property. Consider a variant of the $k$-means
problem that may be regarded as a general version of such problems. Here, the
optimal clusters $O_1, ..., O_k$ are an arbitrary partition of the dataset and
the goal is to output $k$-centers $c_1, ..., c_k$ such that the objective
function $\sum_{i=1}^{k} \sum_{x \in O_{i}} ||x - c_{i}||^2$ is minimized. It
is not difficult to argue that any algorithm (without knowing the optimal
clusters) that outputs a single set of $k$ centers, will not behave well as far
as optimizing the above objective function is concerned. However, this does not
rule out the existence of algorithms that output a list of such $k$ centers
such that at least one of these $k$ centers behaves well. Given an error
parameter $\varepsilon > 0$, let $\ell$ denote the size of the smallest list of
$k$-centers such that at least one of the $k$-centers gives a $(1+\varepsilon)$
approximation w.r.t. the objective function above. In this paper, we show an
upper bound on $\ell$ by giving a randomized algorithm that outputs a list of
$2^{\tilde{O}(k/\varepsilon)}$ $k$-centers. We also give a closely matching
lower bound of $2^{\tilde{\Omega}(k/\sqrt{\varepsilon})}$. Moreover, our
algorithm runs in time $O \left(n d \cdot 2^{\tilde{O}(k/\varepsilon)}
\right)$. This is a significant improvement over the previous result of Ding
and Xu who gave an algorithm with running time $O \left(n d \cdot (\log{n})^{k}
\cdot 2^{poly(k/\varepsilon)} \right)$ and output a list of size $O
\left((\log{n})^k \cdot 2^{poly(k/\varepsilon)} \right)$.
| [
{
"version": "v1",
"created": "Fri, 10 Apr 2015 07:03:58 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Bhattacharya",
"Anup",
""
],
[
"Jaiswal",
"Ragesh",
""
],
[
"Kumar",
"Amit",
""
]
] | TITLE: Faster Algorithms for the Constrained k-means Problem
ABSTRACT: The classical center based clustering problems such as
$k$-means/median/center assume that the optimal clusters satisfy the locality
property that the points in the same cluster are close to each other. A number
of clustering problems arise in machine learning where the optimal clusters do
not follow such a locality property. Consider a variant of the $k$-means
problem that may be regarded as a general version of such problems. Here, the
optimal clusters $O_1, ..., O_k$ are an arbitrary partition of the dataset and
the goal is to output $k$-centers $c_1, ..., c_k$ such that the objective
function $\sum_{i=1}^{k} \sum_{x \in O_{i}} ||x - c_{i}||^2$ is minimized. It
is not difficult to argue that any algorithm (without knowing the optimal
clusters) that outputs a single set of $k$ centers, will not behave well as far
as optimizing the above objective function is concerned. However, this does not
rule out the existence of algorithms that output a list of such $k$ centers
such that at least one of these $k$ centers behaves well. Given an error
parameter $\varepsilon > 0$, let $\ell$ denote the size of the smallest list of
$k$-centers such that at least one of the $k$-centers gives a $(1+\varepsilon)$
approximation w.r.t. the objective function above. In this paper, we show an
upper bound on $\ell$ by giving a randomized algorithm that outputs a list of
$2^{\tilde{O}(k/\varepsilon)}$ $k$-centers. We also give a closely matching
lower bound of $2^{\tilde{\Omega}(k/\sqrt{\varepsilon})}$. Moreover, our
algorithm runs in time $O \left(n d \cdot 2^{\tilde{O}(k/\varepsilon)}
\right)$. This is a significant improvement over the previous result of Ding
and Xu who gave an algorithm with running time $O \left(n d \cdot (\log{n})^{k}
\cdot 2^{poly(k/\varepsilon)} \right)$ and output a list of size $O
\left((\log{n})^k \cdot 2^{poly(k/\varepsilon)} \right)$.
| no_new_dataset | 0.709384 |
1504.02696 | Xinyan Yan | Xinyan Yan, Vadim Indelman, Byron Boots | Incremental Sparse GP Regression for Continuous-time Trajectory
Estimation & Mapping | 10 pages, 10 figures | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent work on simultaneous trajectory estimation and mapping (STEAM) for
mobile robots has found success by representing the trajectory as a Gaussian
process. Gaussian processes can represent a continuous-time trajectory,
elegantly handle asynchronous and sparse measurements, and allow the robot to
query the trajectory to recover its estimated position at any time of interest.
A major drawback of this approach is that STEAM is formulated as a batch
estimation problem. In this paper we provide the critical extensions necessary
to transform the existing batch algorithm into an extremely efficient
incremental algorithm. In particular, we are able to vastly speed up the
solution time through efficient variable reordering and incremental sparse
updates, which we believe will greatly increase the practicality of Gaussian
process methods for robot mapping and localization. Finally, we demonstrate the
approach and its advantages on both synthetic and real datasets.
| [
{
"version": "v1",
"created": "Fri, 10 Apr 2015 14:47:25 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Yan",
"Xinyan",
""
],
[
"Indelman",
"Vadim",
""
],
[
"Boots",
"Byron",
""
]
] | TITLE: Incremental Sparse GP Regression for Continuous-time Trajectory
Estimation & Mapping
ABSTRACT: Recent work on simultaneous trajectory estimation and mapping (STEAM) for
mobile robots has found success by representing the trajectory as a Gaussian
process. Gaussian processes can represent a continuous-time trajectory,
elegantly handle asynchronous and sparse measurements, and allow the robot to
query the trajectory to recover its estimated position at any time of interest.
A major drawback of this approach is that STEAM is formulated as a batch
estimation problem. In this paper we provide the critical extensions necessary
to transform the existing batch algorithm into an extremely efficient
incremental algorithm. In particular, we are able to vastly speed up the
solution time through efficient variable reordering and incremental sparse
updates, which we believe will greatly increase the practicality of Gaussian
process methods for robot mapping and localization. Finally, we demonstrate the
approach and its advantages on both synthetic and real datasets.
| no_new_dataset | 0.709598 |
1504.02764 | Roozbeh Mottaghi | Roozbeh Mottaghi, Yu Xiang, Silvio Savarese | A Coarse-to-Fine Model for 3D Pose Estimation and Sub-category
Recognition | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite the fact that object detection, 3D pose estimation, and sub-category
recognition are highly correlated tasks, they are usually addressed
independently from each other because of the huge space of parameters. To
jointly model all of these tasks, we propose a coarse-to-fine hierarchical
representation, where each level of the hierarchy represents objects at a
different level of granularity. The hierarchical representation prevents
performance loss, which is often caused by the increase in the number of
parameters (as we consider more tasks to model), and the joint modelling
enables resolving ambiguities that exist in independent modelling of these
tasks. We augment PASCAL3D+ dataset with annotations for these tasks and show
that our hierarchical model is effective in joint modelling of object
detection, 3D pose estimation, and sub-category recognition.
| [
{
"version": "v1",
"created": "Fri, 10 Apr 2015 19:18:59 GMT"
}
] | 2015-04-13T00:00:00 | [
[
"Mottaghi",
"Roozbeh",
""
],
[
"Xiang",
"Yu",
""
],
[
"Savarese",
"Silvio",
""
]
] | TITLE: A Coarse-to-Fine Model for 3D Pose Estimation and Sub-category
Recognition
ABSTRACT: Despite the fact that object detection, 3D pose estimation, and sub-category
recognition are highly correlated tasks, they are usually addressed
independently from each other because of the huge space of parameters. To
jointly model all of these tasks, we propose a coarse-to-fine hierarchical
representation, where each level of the hierarchy represents objects at a
different level of granularity. The hierarchical representation prevents
performance loss, which is often caused by the increase in the number of
parameters (as we consider more tasks to model), and the joint modelling
enables resolving ambiguities that exist in independent modelling of these
tasks. We augment PASCAL3D+ dataset with annotations for these tasks and show
that our hierarchical model is effective in joint modelling of object
detection, 3D pose estimation, and sub-category recognition.
| no_new_dataset | 0.709787 |
1406.2989 | Mathias Berglund | Tapani Raiko, Mathias Berglund, Guillaume Alain, Laurent Dinh | Techniques for Learning Binary Stochastic Feedforward Neural Networks | null | null | null | null | stat.ML cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Stochastic binary hidden units in a multi-layer perceptron (MLP) network give
at least three potential benefits when compared to deterministic MLP networks.
(1) They allow to learn one-to-many type of mappings. (2) They can be used in
structured prediction problems, where modeling the internal structure of the
output is important. (3) Stochasticity has been shown to be an excellent
regularizer, which makes generalization performance potentially better in
general. However, training stochastic networks is considerably more difficult.
We study training using M samples of hidden activations per input. We show that
the case M=1 leads to a fundamentally different behavior where the network
tries to avoid stochasticity. We propose two new estimators for the training
gradient and propose benchmark tests for comparing training algorithms. Our
experiments confirm that training stochastic networks is difficult and show
that the proposed two estimators perform favorably among all the five known
estimators.
| [
{
"version": "v1",
"created": "Wed, 11 Jun 2014 18:29:27 GMT"
},
{
"version": "v2",
"created": "Fri, 19 Dec 2014 13:37:05 GMT"
},
{
"version": "v3",
"created": "Thu, 9 Apr 2015 12:58:06 GMT"
}
] | 2015-04-10T00:00:00 | [
[
"Raiko",
"Tapani",
""
],
[
"Berglund",
"Mathias",
""
],
[
"Alain",
"Guillaume",
""
],
[
"Dinh",
"Laurent",
""
]
] | TITLE: Techniques for Learning Binary Stochastic Feedforward Neural Networks
ABSTRACT: Stochastic binary hidden units in a multi-layer perceptron (MLP) network give
at least three potential benefits when compared to deterministic MLP networks.
(1) They allow to learn one-to-many type of mappings. (2) They can be used in
structured prediction problems, where modeling the internal structure of the
output is important. (3) Stochasticity has been shown to be an excellent
regularizer, which makes generalization performance potentially better in
general. However, training stochastic networks is considerably more difficult.
We study training using M samples of hidden activations per input. We show that
the case M=1 leads to a fundamentally different behavior where the network
tries to avoid stochasticity. We propose two new estimators for the training
gradient and propose benchmark tests for comparing training algorithms. Our
experiments confirm that training stochastic networks is difficult and show
that the proposed two estimators perform favorably among all the five known
estimators.
| no_new_dataset | 0.711017 |
1408.1054 | Wojciech Czarnecki | Wojciech Marian Czarnecki, Jacek Tabor | Multithreshold Entropy Linear Classifier | null | null | 10.1016/j.eswa.2015.03.007 | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Linear classifiers separate the data with a hyperplane. In this paper we
focus on the novel method of construction of multithreshold linear classifier,
which separates the data with multiple parallel hyperplanes. Proposed model is
based on the information theory concepts -- namely Renyi's quadratic entropy
and Cauchy-Schwarz divergence.
We begin with some general properties, including data scale invariance. Then
we prove that our method is a multithreshold large margin classifier, which
shows the analogy to the SVM, while in the same time works with much broader
class of hypotheses. What is also interesting, proposed method is aimed at the
maximization of the balanced quality measure (such as Matthew's Correlation
Coefficient) as opposed to very common maximization of the accuracy. This
feature comes directly from the optimization problem statement and is further
confirmed by the experiments on the UCI datasets.
It appears, that our Multithreshold Entropy Linear Classifier (MELC) obtaines
similar or higher scores than the ones given by SVM on both synthetic and real
data. We show how proposed approach can be benefitial for the cheminformatics
in the task of ligands activity prediction, where despite better classification
results, MELC gives some additional insight into the data structure (classes of
underrepresented chemical compunds).
| [
{
"version": "v1",
"created": "Mon, 4 Aug 2014 18:01:29 GMT"
}
] | 2015-04-10T00:00:00 | [
[
"Czarnecki",
"Wojciech Marian",
""
],
[
"Tabor",
"Jacek",
""
]
] | TITLE: Multithreshold Entropy Linear Classifier
ABSTRACT: Linear classifiers separate the data with a hyperplane. In this paper we
focus on the novel method of construction of multithreshold linear classifier,
which separates the data with multiple parallel hyperplanes. Proposed model is
based on the information theory concepts -- namely Renyi's quadratic entropy
and Cauchy-Schwarz divergence.
We begin with some general properties, including data scale invariance. Then
we prove that our method is a multithreshold large margin classifier, which
shows the analogy to the SVM, while in the same time works with much broader
class of hypotheses. What is also interesting, proposed method is aimed at the
maximization of the balanced quality measure (such as Matthew's Correlation
Coefficient) as opposed to very common maximization of the accuracy. This
feature comes directly from the optimization problem statement and is further
confirmed by the experiments on the UCI datasets.
It appears, that our Multithreshold Entropy Linear Classifier (MELC) obtaines
similar or higher scores than the ones given by SVM on both synthetic and real
data. We show how proposed approach can be benefitial for the cheminformatics
in the task of ligands activity prediction, where despite better classification
results, MELC gives some additional insight into the data structure (classes of
underrepresented chemical compunds).
| no_new_dataset | 0.710478 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.