id
stringlengths 9
16
| submitter
stringlengths 3
64
⌀ | authors
stringlengths 5
6.63k
| title
stringlengths 7
245
| comments
stringlengths 1
482
⌀ | journal-ref
stringlengths 4
382
⌀ | doi
stringlengths 9
151
⌀ | report-no
stringclasses 984
values | categories
stringlengths 5
108
| license
stringclasses 9
values | abstract
stringlengths 83
3.41k
| versions
listlengths 1
20
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-04-11 00:00:00
| authors_parsed
sequencelengths 1
427
| prompt
stringlengths 166
3.49k
| label
stringclasses 2
values | prob
float64 0.5
0.98
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1603.03656 | Nick Feamster | Nick Feamster | Revealing Utilization at Internet Interconnection Points | null | null | null | null | cs.NI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent Internet interconnection disputes have sparked an in- creased interest
in developing methods for gathering and collecting data about utilization at
interconnection points. One mechanism, developed by DeepField Networks, allows
Internet service providers (ISPs) to gather and aggregate utilization
information using network flow statistics, standardized in the Internet
Engineering Task Force as IPFIX. This report (1) provides an overview of the
method that DeepField Networks is using to measure the utilization of various
interconnection links between content providers and ISPs or links over which
traffic between content and ISPs flow; and (2) surveys the findings from five
months of Internet utilization data provided by seven participating
ISPs---Bright House Networks, Comcast, Cox, Mediacom, Midco, Suddenlink, and
Time Warner Cable---whose access networks represent about 50% of all U.S.
broadband subscribers. The dataset includes about 97% of the paid peering,
settlement-free peering, and ISP-paid transit links of each of the
participating ISPs. Initial analysis of the data---which comprises more than
1,000 link groups, representing the diverse and substitutable available
routes---suggests that many interconnects have significant spare capacity, that
this spare capacity exists both across ISPs in each region and in aggregate for
any individual ISP, and that the aggregate utilization across interconnects
interconnects is roughly 50% during peak periods.
| [
{
"version": "v1",
"created": "Fri, 11 Mar 2016 15:06:18 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Sep 2016 01:10:58 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Feamster",
"Nick",
""
]
] | TITLE: Revealing Utilization at Internet Interconnection Points
ABSTRACT: Recent Internet interconnection disputes have sparked an in- creased interest
in developing methods for gathering and collecting data about utilization at
interconnection points. One mechanism, developed by DeepField Networks, allows
Internet service providers (ISPs) to gather and aggregate utilization
information using network flow statistics, standardized in the Internet
Engineering Task Force as IPFIX. This report (1) provides an overview of the
method that DeepField Networks is using to measure the utilization of various
interconnection links between content providers and ISPs or links over which
traffic between content and ISPs flow; and (2) surveys the findings from five
months of Internet utilization data provided by seven participating
ISPs---Bright House Networks, Comcast, Cox, Mediacom, Midco, Suddenlink, and
Time Warner Cable---whose access networks represent about 50% of all U.S.
broadband subscribers. The dataset includes about 97% of the paid peering,
settlement-free peering, and ISP-paid transit links of each of the
participating ISPs. Initial analysis of the data---which comprises more than
1,000 link groups, representing the diverse and substitutable available
routes---suggests that many interconnects have significant spare capacity, that
this spare capacity exists both across ISPs in each region and in aggregate for
any individual ISP, and that the aggregate utilization across interconnects
interconnects is roughly 50% during peak periods.
| no_new_dataset | 0.933915 |
1607.07959 | Ansaf Salleb-Aouissi | Ilia Vovsha, Ansaf Salleb-Aouissi, Anita Raja, Thomas Koch, Alex
Rybchuk, Axinia Radeva, Ashwath Rajan, Yiwen Huang, Hatim Diab, Ashish Tomar,
and Ronald Wapner | Using Kernel Methods and Model Selection for Prediction of Preterm Birth | Presented at 2016 Machine Learning and Healthcare Conference (MLHC
2016), Los Angeles, CA. In this revision, we updated page 4 by adding the
reference Vovsha et al. (2013) (incorrectly referenced as XXX in the previous
version due to double blind reviewing). The bibtex entry is now added to the
references | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We describe an application of machine learning to the problem of predicting
preterm birth. We conduct a secondary analysis on a clinical trial dataset
collected by the National In- stitute of Child Health and Human Development
(NICHD) while focusing our attention on predicting different classes of preterm
birth. We compare three approaches for deriving predictive models: a support
vector machine (SVM) approach with linear and non-linear kernels, logistic
regression with different model selection along with a model based on decision
rules prescribed by physician experts for prediction of preterm birth. Our
approach highlights the pre-processing methods applied to handle the inherent
dynamics, noise and gaps in the data and describe techniques used to handle
skewed class distributions. Empirical experiments demonstrate significant
improvement in predicting preterm birth compared to past work.
| [
{
"version": "v1",
"created": "Wed, 27 Jul 2016 04:56:57 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Sep 2016 12:25:00 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Vovsha",
"Ilia",
""
],
[
"Salleb-Aouissi",
"Ansaf",
""
],
[
"Raja",
"Anita",
""
],
[
"Koch",
"Thomas",
""
],
[
"Rybchuk",
"Alex",
""
],
[
"Radeva",
"Axinia",
""
],
[
"Rajan",
"Ashwath",
""
],
[
"Huang",
"Yiwen",
""
],
[
"Diab",
"Hatim",
""
],
[
"Tomar",
"Ashish",
""
],
[
"Wapner",
"Ronald",
""
]
] | TITLE: Using Kernel Methods and Model Selection for Prediction of Preterm Birth
ABSTRACT: We describe an application of machine learning to the problem of predicting
preterm birth. We conduct a secondary analysis on a clinical trial dataset
collected by the National In- stitute of Child Health and Human Development
(NICHD) while focusing our attention on predicting different classes of preterm
birth. We compare three approaches for deriving predictive models: a support
vector machine (SVM) approach with linear and non-linear kernels, logistic
regression with different model selection along with a model based on decision
rules prescribed by physician experts for prediction of preterm birth. Our
approach highlights the pre-processing methods applied to handle the inherent
dynamics, noise and gaps in the data and describe techniques used to handle
skewed class distributions. Empirical experiments demonstrate significant
improvement in predicting preterm birth compared to past work.
| no_new_dataset | 0.945096 |
1608.02875 | Haishan Ye | Haishan Ye, Luo Luo and Zhihua Zhang | Revisiting Sub-sampled Newton Methods | null | null | null | null | math.OC cs.NA | http://creativecommons.org/licenses/by/4.0/ | Many machine learning models depend on solving a large scale optimization
problem. Recently, sub-sampled Newton methods have emerged to attract much
attention for optimization due to their efficiency at each iteration, rectified
a weakness in the ordinary Newton method of suffering a high cost at each
iteration while commanding a high convergence rate. In this work we propose two
new efficient Newton-type methods, Refined Sub-sampled Newton and Refined
Sketch Newton. Our methods exhibit a great advantage over existing sub-sampled
Newton methods, especially when Hessian-vector multiplication can be calculated
efficiently. Specifically, the proposed methods are shown to converge
superlinearly in general case and quadratically under a little stronger
assumption. The proposed methods can be generalized to a unifying framework for
the convergence proof of several existing sub-sampled Newton methods, revealing
new convergence properties. Finally, we empirically evaluate the performance of
our methods on several standard datasets and the results show consistent
improvement in computational efficiency.
| [
{
"version": "v1",
"created": "Mon, 8 Aug 2016 03:30:12 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Sep 2016 03:57:35 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Ye",
"Haishan",
""
],
[
"Luo",
"Luo",
""
],
[
"Zhang",
"Zhihua",
""
]
] | TITLE: Revisiting Sub-sampled Newton Methods
ABSTRACT: Many machine learning models depend on solving a large scale optimization
problem. Recently, sub-sampled Newton methods have emerged to attract much
attention for optimization due to their efficiency at each iteration, rectified
a weakness in the ordinary Newton method of suffering a high cost at each
iteration while commanding a high convergence rate. In this work we propose two
new efficient Newton-type methods, Refined Sub-sampled Newton and Refined
Sketch Newton. Our methods exhibit a great advantage over existing sub-sampled
Newton methods, especially when Hessian-vector multiplication can be calculated
efficiently. Specifically, the proposed methods are shown to converge
superlinearly in general case and quadratically under a little stronger
assumption. The proposed methods can be generalized to a unifying framework for
the convergence proof of several existing sub-sampled Newton methods, revealing
new convergence properties. Finally, we empirically evaluate the performance of
our methods on several standard datasets and the results show consistent
improvement in computational efficiency.
| no_new_dataset | 0.947817 |
1608.05921 | Dongwoo Kim | Dongwoo Kim, Lexing Xie, Cheng Soon Ong | Probabilistic Knowledge Graph Construction: Compositional and
Incremental Approaches | The 25th ACM International Conference on Information and Knowledge
Management (CIKM 2016) | null | 10.1145/2983323.2983677 | null | stat.ML cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Knowledge graph construction consists of two tasks: extracting information
from external resources (knowledge population) and inferring missing
information through a statistical analysis on the extracted information
(knowledge completion). In many cases, insufficient external resources in the
knowledge population hinder the subsequent statistical inference. The gap
between these two processes can be reduced by an incremental population
approach. We propose a new probabilistic knowledge graph factorisation method
that benefits from the path structure of existing knowledge (e.g. syllogism)
and enables a common modelling approach to be used for both incremental
population and knowledge completion tasks. More specifically, the probabilistic
formulation allows us to develop an incremental population algorithm that
trades off exploitation-exploration. Experiments on three benchmark datasets
show that the balanced exploitation-exploration helps the incremental
population, and the additional path structure helps to predict missing
information in knowledge completion.
| [
{
"version": "v1",
"created": "Sun, 21 Aug 2016 11:49:53 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Sep 2016 04:52:33 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Kim",
"Dongwoo",
""
],
[
"Xie",
"Lexing",
""
],
[
"Ong",
"Cheng Soon",
""
]
] | TITLE: Probabilistic Knowledge Graph Construction: Compositional and
Incremental Approaches
ABSTRACT: Knowledge graph construction consists of two tasks: extracting information
from external resources (knowledge population) and inferring missing
information through a statistical analysis on the extracted information
(knowledge completion). In many cases, insufficient external resources in the
knowledge population hinder the subsequent statistical inference. The gap
between these two processes can be reduced by an incremental population
approach. We propose a new probabilistic knowledge graph factorisation method
that benefits from the path structure of existing knowledge (e.g. syllogism)
and enables a common modelling approach to be used for both incremental
population and knowledge completion tasks. More specifically, the probabilistic
formulation allows us to develop an incremental population algorithm that
trades off exploitation-exploration. Experiments on three benchmark datasets
show that the balanced exploitation-exploration helps the incremental
population, and the additional path structure helps to predict missing
information in knowledge completion.
| no_new_dataset | 0.949153 |
1608.08435 | Rajasekar Venkatesan | Rajasekar Venkatesan, Meng Joo Er | Multi-Label Classification Method Based on Extreme Learning Machines | 6 pages, 7 figures, 7 tables, ICARCV | null | 10.1109/ICARCV.2014.7064375 | null | cs.LG cs.AI cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, an Extreme Learning Machine (ELM) based technique for
Multi-label classification problems is proposed and discussed. In multi-label
classification, each of the input data samples belongs to one or more than one
class labels. The traditional binary and multi-class classification problems
are the subset of the multi-label problem with the number of labels
corresponding to each sample limited to one. The proposed ELM based multi-label
classification technique is evaluated with six different benchmark multi-label
datasets from different domains such as multimedia, text and biology. A
detailed comparison of the results is made by comparing the proposed method
with the results from nine state of the arts techniques for five different
evaluation metrics. The nine methods are chosen from different categories of
multi-label methods. The comparative results shows that the proposed Extreme
Learning Machine based multi-label classification technique is a better
alternative than the existing state of the art methods for multi-label
problems.
| [
{
"version": "v1",
"created": "Tue, 30 Aug 2016 13:08:06 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Venkatesan",
"Rajasekar",
""
],
[
"Er",
"Meng Joo",
""
]
] | TITLE: Multi-Label Classification Method Based on Extreme Learning Machines
ABSTRACT: In this paper, an Extreme Learning Machine (ELM) based technique for
Multi-label classification problems is proposed and discussed. In multi-label
classification, each of the input data samples belongs to one or more than one
class labels. The traditional binary and multi-class classification problems
are the subset of the multi-label problem with the number of labels
corresponding to each sample limited to one. The proposed ELM based multi-label
classification technique is evaluated with six different benchmark multi-label
datasets from different domains such as multimedia, text and biology. A
detailed comparison of the results is made by comparing the proposed method
with the results from nine state of the arts techniques for five different
evaluation metrics. The nine methods are chosen from different categories of
multi-label methods. The comparative results shows that the proposed Extreme
Learning Machine based multi-label classification technique is a better
alternative than the existing state of the art methods for multi-label
problems.
| no_new_dataset | 0.949623 |
1608.08898 | Rajasekar Venkatesan | Meng Joo Er, Rajasekar Venkatesan and Ning Wang | A High Speed Multi-label Classifier based on Extreme Learning Machines | 12 pages, 2 figures, 10 tables | null | 10.1007/978-3-319-28373-9_37 | null | cs.LG cs.AI cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper a high speed neural network classifier based on extreme
learning machines for multi-label classification problem is proposed and
dis-cussed. Multi-label classification is a superset of traditional binary and
multi-class classification problems. The proposed work extends the extreme
learning machine technique to adapt to the multi-label problems. As opposed to
the single-label problem, both the number of labels the sample belongs to, and
each of those target labels are to be identified for multi-label classification
resulting in in-creased complexity. The proposed high speed multi-label
classifier is applied to six benchmark datasets comprising of different
application areas such as multi-media, text and biology. The training time and
testing time of the classifier are compared with those of the state-of-the-arts
methods. Experimental studies show that for all the six datasets, our proposed
technique have faster execution speed and better performance, thereby
outperforming all the existing multi-label clas-sification methods.
| [
{
"version": "v1",
"created": "Wed, 31 Aug 2016 14:56:12 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Er",
"Meng Joo",
""
],
[
"Venkatesan",
"Rajasekar",
""
],
[
"Wang",
"Ning",
""
]
] | TITLE: A High Speed Multi-label Classifier based on Extreme Learning Machines
ABSTRACT: In this paper a high speed neural network classifier based on extreme
learning machines for multi-label classification problem is proposed and
dis-cussed. Multi-label classification is a superset of traditional binary and
multi-class classification problems. The proposed work extends the extreme
learning machine technique to adapt to the multi-label problems. As opposed to
the single-label problem, both the number of labels the sample belongs to, and
each of those target labels are to be identified for multi-label classification
resulting in in-creased complexity. The proposed high speed multi-label
classifier is applied to six benchmark datasets comprising of different
application areas such as multi-media, text and biology. The training time and
testing time of the classifier are compared with those of the state-of-the-arts
methods. Experimental studies show that for all the six datasets, our proposed
technique have faster execution speed and better performance, thereby
outperforming all the existing multi-label clas-sification methods.
| no_new_dataset | 0.947721 |
1609.00086 | Rajasekar Venkatesan | Rajasekar Venkatesan, Meng Joo Er, Mihika Dave, Mahardhika Pratama,
Shiqian Wu | A novel online multi-label classifier for high-speed streaming data
applications | 18 pages, 7 tables, 3 figures. arXiv admin note: text overlap with
arXiv:1608.08898 | null | 10.1007/s12530-016-9162-8 | null | cs.LG cs.AI cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, a high-speed online neural network classifier based on extreme
learning machines for multi-label classification is proposed. In multi-label
classification, each of the input data sample belongs to one or more than one
of the target labels. The traditional binary and multi-class classification
where each sample belongs to only one target class forms the subset of
multi-label classification. Multi-label classification problems are far more
complex than binary and multi-class classification problems, as both the number
of target labels and each of the target labels corresponding to each of the
input samples are to be identified. The proposed work exploits the high-speed
nature of the extreme learning machines to achieve real-time multi-label
classification of streaming data. A new threshold-based online sequential
learning algorithm is proposed for high speed and streaming data classification
of multi-label problems. The proposed method is experimented with six different
datasets from different application domains such as multimedia, text, and
biology. The hamming loss, accuracy, training time and testing time of the
proposed technique is compared with nine different state-of-the-art methods.
Experimental studies shows that the proposed technique outperforms the existing
multi-label classifiers in terms of performance and speed.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 01:58:50 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Venkatesan",
"Rajasekar",
""
],
[
"Er",
"Meng Joo",
""
],
[
"Dave",
"Mihika",
""
],
[
"Pratama",
"Mahardhika",
""
],
[
"Wu",
"Shiqian",
""
]
] | TITLE: A novel online multi-label classifier for high-speed streaming data
applications
ABSTRACT: In this paper, a high-speed online neural network classifier based on extreme
learning machines for multi-label classification is proposed. In multi-label
classification, each of the input data sample belongs to one or more than one
of the target labels. The traditional binary and multi-class classification
where each sample belongs to only one target class forms the subset of
multi-label classification. Multi-label classification problems are far more
complex than binary and multi-class classification problems, as both the number
of target labels and each of the target labels corresponding to each of the
input samples are to be identified. The proposed work exploits the high-speed
nature of the extreme learning machines to achieve real-time multi-label
classification of streaming data. A new threshold-based online sequential
learning algorithm is proposed for high speed and streaming data classification
of multi-label problems. The proposed method is experimented with six different
datasets from different application domains such as multimedia, text, and
biology. The hamming loss, accuracy, training time and testing time of the
proposed technique is compared with nine different state-of-the-art methods.
Experimental studies shows that the proposed technique outperforms the existing
multi-label classifiers in terms of performance and speed.
| no_new_dataset | 0.948537 |
1609.00435 | David Jurgens | David Jurgens, Srijan Kumar, Raine Hoover, Dan McFarland, Dan Jurafsky | Citation Classification for Behavioral Analysis of a Scientific Field | null | null | null | null | cs.CL cs.DL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Citations are an important indicator of the state of a scientific field,
reflecting how authors frame their work, and influencing uptake by future
scholars. However, our understanding of citation behavior has been limited to
small-scale manual citation analysis. We perform the largest behavioral study
of citations to date, analyzing how citations are both framed and taken up by
scholars in one entire field: natural language processing. We introduce a new
dataset of nearly 2,000 citations annotated for function and centrality, and
use it to develop a state-of-the-art classifier and label the entire ACL
Reference Corpus. We then study how citations are framed by authors and use
both papers and online traces to track how citations are followed by readers.
We demonstrate that authors are sensitive to discourse structure and
publication venue when citing, that online readers follow temporal links to
previous and future work rather than methodological links, and that how a paper
cites related work is predictive of its citation count. Finally, we use changes
in citation roles to show that the field of NLP is undergoing a significant
increase in consensus.
| [
{
"version": "v1",
"created": "Fri, 2 Sep 2016 00:40:15 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Jurgens",
"David",
""
],
[
"Kumar",
"Srijan",
""
],
[
"Hoover",
"Raine",
""
],
[
"McFarland",
"Dan",
""
],
[
"Jurafsky",
"Dan",
""
]
] | TITLE: Citation Classification for Behavioral Analysis of a Scientific Field
ABSTRACT: Citations are an important indicator of the state of a scientific field,
reflecting how authors frame their work, and influencing uptake by future
scholars. However, our understanding of citation behavior has been limited to
small-scale manual citation analysis. We perform the largest behavioral study
of citations to date, analyzing how citations are both framed and taken up by
scholars in one entire field: natural language processing. We introduce a new
dataset of nearly 2,000 citations annotated for function and centrality, and
use it to develop a state-of-the-art classifier and label the entire ACL
Reference Corpus. We then study how citations are framed by authors and use
both papers and online traces to track how citations are followed by readers.
We demonstrate that authors are sensitive to discourse structure and
publication venue when citing, that online readers follow temporal links to
previous and future work rather than methodological links, and that how a paper
cites related work is predictive of its citation count. Finally, we use changes
in citation roles to show that the field of NLP is undergoing a significant
increase in consensus.
| new_dataset | 0.965152 |
1609.00843 | Rajasekar Venkatesan | Meng Joo Er, Rajasekar Venkatesan, Ning Wang | An Online Universal Classifier for Binary, Multi-class and Multi-label
Classification | 6 pages, 6 tables | null | null | null | cs.LG cs.AI cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Classification involves the learning of the mapping function that associates
input samples to corresponding target label. There are two major categories of
classification problems: Single-label classification and Multi-label
classification. Traditional binary and multi-class classifications are
sub-categories of single-label classification. Several classifiers are
developed for binary, multi-class and multi-label classification problems, but
there are no classifiers available in the literature capable of performing all
three types of classification. In this paper, a novel online universal
classifier capable of performing all the three types of classification is
proposed. Being a high speed online classifier, the proposed technique can be
applied to streaming data applications. The performance of the developed
classifier is evaluated using datasets from binary, multi-class and multi-label
problems. The results obtained are compared with state-of-the-art techniques
from each of the classification types.
| [
{
"version": "v1",
"created": "Sat, 3 Sep 2016 17:03:14 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Er",
"Meng Joo",
""
],
[
"Venkatesan",
"Rajasekar",
""
],
[
"Wang",
"Ning",
""
]
] | TITLE: An Online Universal Classifier for Binary, Multi-class and Multi-label
Classification
ABSTRACT: Classification involves the learning of the mapping function that associates
input samples to corresponding target label. There are two major categories of
classification problems: Single-label classification and Multi-label
classification. Traditional binary and multi-class classifications are
sub-categories of single-label classification. Several classifiers are
developed for binary, multi-class and multi-label classification problems, but
there are no classifiers available in the literature capable of performing all
three types of classification. In this paper, a novel online universal
classifier capable of performing all the three types of classification is
proposed. Being a high speed online classifier, the proposed technique can be
applied to streaming data applications. The performance of the developed
classifier is evaluated using datasets from binary, multi-class and multi-label
problems. The results obtained are compared with state-of-the-art techniques
from each of the classification types.
| no_new_dataset | 0.946001 |
1609.00845 | Kwang-Sung Jun | Kwang-Sung Jun and Robert Nowak | Graph-Based Active Learning: A New Look at Expected Error Minimization | Submitted to GlobalSIP 2016 | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In graph-based active learning, algorithms based on expected error
minimization (EEM) have been popular and yield good empirical performance. The
exact computation of EEM optimally balances exploration and exploitation. In
practice, however, EEM-based algorithms employ various approximations due to
the computational hardness of exact EEM. This can result in a lack of either
exploration or exploitation, which can negatively impact the effectiveness of
active learning. We propose a new algorithm TSA (Two-Step Approximation) that
balances between exploration and exploitation efficiently while enjoying the
same computational complexity as existing approximations. Finally, we
empirically show the value of balancing between exploration and exploitation in
both toy and real-world datasets where our method outperforms several
state-of-the-art methods.
| [
{
"version": "v1",
"created": "Sat, 3 Sep 2016 17:30:15 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Jun",
"Kwang-Sung",
""
],
[
"Nowak",
"Robert",
""
]
] | TITLE: Graph-Based Active Learning: A New Look at Expected Error Minimization
ABSTRACT: In graph-based active learning, algorithms based on expected error
minimization (EEM) have been popular and yield good empirical performance. The
exact computation of EEM optimally balances exploration and exploitation. In
practice, however, EEM-based algorithms employ various approximations due to
the computational hardness of exact EEM. This can result in a lack of either
exploration or exploitation, which can negatively impact the effectiveness of
active learning. We propose a new algorithm TSA (Two-Step Approximation) that
balances between exploration and exploitation efficiently while enjoying the
same computational complexity as existing approximations. Finally, we
empirically show the value of balancing between exploration and exploitation in
both toy and real-world datasets where our method outperforms several
state-of-the-art methods.
| no_new_dataset | 0.944536 |
1609.00878 | Joao Papa | Silas E. N. Fernandes, Danillo R. Pereira, Caio C. O. Ramos, Andre N.
Souza and Joao P. Papa | A Probabilistic Optimum-Path Forest Classifier for Binary Classification
Problems | Submitted to Neural Processing Letters | null | null | null | cs.CV cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Probabilistic-driven classification techniques extend the role of traditional
approaches that output labels (usually integer numbers) only. Such techniques
are more fruitful when dealing with problems where one is not interested in
recognition/identification only, but also into monitoring the behavior of
consumers and/or machines, for instance. Therefore, by means of probability
estimates, one can take decisions to work better in a number of scenarios. In
this paper, we propose a probabilistic-based Optimum Path Forest (OPF)
classifier to handle with binary classification problems, and we show it can be
more accurate than naive OPF in a number of datasets. In addition to being just
more accurate or not, probabilistic OPF turns to be another useful tool to the
scientific community.
| [
{
"version": "v1",
"created": "Sun, 4 Sep 2016 00:12:04 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Fernandes",
"Silas E. N.",
""
],
[
"Pereira",
"Danillo R.",
""
],
[
"Ramos",
"Caio C. O.",
""
],
[
"Souza",
"Andre N.",
""
],
[
"Papa",
"Joao P.",
""
]
] | TITLE: A Probabilistic Optimum-Path Forest Classifier for Binary Classification
Problems
ABSTRACT: Probabilistic-driven classification techniques extend the role of traditional
approaches that output labels (usually integer numbers) only. Such techniques
are more fruitful when dealing with problems where one is not interested in
recognition/identification only, but also into monitoring the behavior of
consumers and/or machines, for instance. Therefore, by means of probability
estimates, one can take decisions to work better in a number of scenarios. In
this paper, we propose a probabilistic-based Optimum Path Forest (OPF)
classifier to handle with binary classification problems, and we show it can be
more accurate than naive OPF in a number of datasets. In addition to being just
more accurate or not, probabilistic OPF turns to be another useful tool to the
scientific community.
| no_new_dataset | 0.950503 |
1609.00904 | Eric Holloway | Eric Holloway and Robert Marks II | High Dimensional Human Guided Machine Learning | 3 pages, 1 figure, HCOMP 2016 submission, work in progress | null | null | null | cs.AI cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Have you ever looked at a machine learning classification model and thought,
I could have made that? Well, that is what we test in this project, comparing
XGBoost trained on human engineered features to training directly on data. The
human engineered features do not outperform XGBoost trained di- rectly on the
data, but they are comparable. This project con- tributes a novel method for
utilizing human created classifi- cation models on high dimensional datasets.
| [
{
"version": "v1",
"created": "Sun, 4 Sep 2016 08:45:26 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Holloway",
"Eric",
""
],
[
"Marks",
"Robert",
"II"
]
] | TITLE: High Dimensional Human Guided Machine Learning
ABSTRACT: Have you ever looked at a machine learning classification model and thought,
I could have made that? Well, that is what we test in this project, comparing
XGBoost trained on human engineered features to training directly on data. The
human engineered features do not outperform XGBoost trained di- rectly on the
data, but they are comparable. This project con- tributes a novel method for
utilizing human created classifi- cation models on high dimensional datasets.
| no_new_dataset | 0.948298 |
1609.00921 | Muhammad Yousefnezhad | Muhammad Yousefnezhad and Daoqiang Zhang | Decoding visual stimuli in human brain by using Anatomical Pattern
Analysis on fMRI images | The 8th International Conference on Brain Inspired Cognitive Systems
(BICS'16), Beijing, China, Nov/28-30/2016 | null | null | null | stat.ML cs.LG q-bio.NC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A universal unanswered question in neuroscience and machine learning is
whether computers can decode the patterns of the human brain. Multi-Voxels
Pattern Analysis (MVPA) is a critical tool for addressing this question.
However, there are two challenges in the previous MVPA methods, which include
decreasing sparsity and noises in the extracted features and increasing the
performance of prediction. In overcoming mentioned challenges, this paper
proposes Anatomical Pattern Analysis (APA) for decoding visual stimuli in the
human brain. This framework develops a novel anatomical feature extraction
method and a new imbalance AdaBoost algorithm for binary classification.
Further, it utilizes an Error-Correcting Output Codes (ECOC) method for
multi-class prediction. APA can automatically detect active regions for each
category of the visual stimuli. Moreover, it enables us to combine homogeneous
datasets for applying advanced classification. Experimental studies on 4 visual
categories (words, consonants, objects and scrambled photos) demonstrate that
the proposed approach achieves superior performance to state-of-the-art
methods.
| [
{
"version": "v1",
"created": "Sun, 4 Sep 2016 12:01:50 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Yousefnezhad",
"Muhammad",
""
],
[
"Zhang",
"Daoqiang",
""
]
] | TITLE: Decoding visual stimuli in human brain by using Anatomical Pattern
Analysis on fMRI images
ABSTRACT: A universal unanswered question in neuroscience and machine learning is
whether computers can decode the patterns of the human brain. Multi-Voxels
Pattern Analysis (MVPA) is a critical tool for addressing this question.
However, there are two challenges in the previous MVPA methods, which include
decreasing sparsity and noises in the extracted features and increasing the
performance of prediction. In overcoming mentioned challenges, this paper
proposes Anatomical Pattern Analysis (APA) for decoding visual stimuli in the
human brain. This framework develops a novel anatomical feature extraction
method and a new imbalance AdaBoost algorithm for binary classification.
Further, it utilizes an Error-Correcting Output Codes (ECOC) method for
multi-class prediction. APA can automatically detect active regions for each
category of the visual stimuli. Moreover, it enables us to combine homogeneous
datasets for applying advanced classification. Experimental studies on 4 visual
categories (words, consonants, objects and scrambled photos) demonstrate that
the proposed approach achieves superior performance to state-of-the-art
methods.
| no_new_dataset | 0.946843 |
1609.00969 | Christina Lioma Assoc. Prof | Casper Petersen and Jakob Grue Simonsen and Kalervo Jarvelin and
Christina Lioma | Adaptive Distributional Extensions to DFR Ranking | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Divergence From Randomness (DFR) ranking models assume that informative terms
are distributed in a corpus differently than non-informative terms. Different
statistical models (e.g. Poisson, geometric) are used to model the distribution
of non-informative terms, producing different DFR models. An informative term
is then detected by measuring the divergence of its distribution from the
distribution of non-informative terms. However, there is little empirical
evidence that the distributions of non-informative terms used in DFR actually
fit current datasets. Practically this risks providing a poor separation
between informative and non-informative terms, thus compromising the
discriminative power of the ranking model. We present a novel extension to DFR,
which first detects the best-fitting distribution of non-informative terms in a
collection, and then adapts the ranking computation to this best-fitting
distribution. We call this model Adaptive Distributional Ranking (ADR) because
it adapts the ranking to the statistics of the specific dataset being processed
each time. Experiments on TREC data show ADR to outperform DFR models (and
their extensions) and be comparable in performance to a query likelihood
language model (LM).
| [
{
"version": "v1",
"created": "Sun, 4 Sep 2016 17:55:39 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Petersen",
"Casper",
""
],
[
"Simonsen",
"Jakob Grue",
""
],
[
"Jarvelin",
"Kalervo",
""
],
[
"Lioma",
"Christina",
""
]
] | TITLE: Adaptive Distributional Extensions to DFR Ranking
ABSTRACT: Divergence From Randomness (DFR) ranking models assume that informative terms
are distributed in a corpus differently than non-informative terms. Different
statistical models (e.g. Poisson, geometric) are used to model the distribution
of non-informative terms, producing different DFR models. An informative term
is then detected by measuring the divergence of its distribution from the
distribution of non-informative terms. However, there is little empirical
evidence that the distributions of non-informative terms used in DFR actually
fit current datasets. Practically this risks providing a poor separation
between informative and non-informative terms, thus compromising the
discriminative power of the ranking model. We present a novel extension to DFR,
which first detects the best-fitting distribution of non-informative terms in a
collection, and then adapts the ranking computation to this best-fitting
distribution. We call this model Adaptive Distributional Ranking (ADR) because
it adapts the ranking to the statistics of the specific dataset being processed
each time. Experiments on TREC data show ADR to outperform DFR models (and
their extensions) and be comparable in performance to a query likelihood
language model (LM).
| no_new_dataset | 0.946349 |
1609.00990 | Nhien-An Le-Khac | Nhien-An Le-Khac, Sammer Markos, Tahar Kechadi | A data mining-based solution for detecting suspicious money laundering
cases in an investment bank | null | null | null | null | cs.DB cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Today, money laundering poses a serious threat not only to financial
institutions but also to the nation. This criminal activity is becoming more
and more sophisticated and seems to have moved from the clichy of drug
trafficking to financing terrorism and surely not forgetting personal gain.
Most international financial institutions have been implementing anti-money
laundering solutions to fight investment fraud. However, traditional
investigative techniques consume numerous man-hours. Recently, data mining
approaches have been developed and are considered as well-suited techniques for
detecting money laundering activities. Within the scope of a collaboration
project for the purpose of developing a new solution for the anti-money
laundering Units in an international investment bank, we proposed a simple and
efficient data mining-based solution for anti-money laundering. In this paper,
we present this solution developed as a tool and show some preliminary
experiment results with real transaction datasets.
| [
{
"version": "v1",
"created": "Sun, 4 Sep 2016 21:03:32 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Le-Khac",
"Nhien-An",
""
],
[
"Markos",
"Sammer",
""
],
[
"Kechadi",
"Tahar",
""
]
] | TITLE: A data mining-based solution for detecting suspicious money laundering
cases in an investment bank
ABSTRACT: Today, money laundering poses a serious threat not only to financial
institutions but also to the nation. This criminal activity is becoming more
and more sophisticated and seems to have moved from the clichy of drug
trafficking to financing terrorism and surely not forgetting personal gain.
Most international financial institutions have been implementing anti-money
laundering solutions to fight investment fraud. However, traditional
investigative techniques consume numerous man-hours. Recently, data mining
approaches have been developed and are considered as well-suited techniques for
detecting money laundering activities. Within the scope of a collaboration
project for the purpose of developing a new solution for the anti-money
laundering Units in an international investment bank, we proposed a simple and
efficient data mining-based solution for anti-money laundering. In this paper,
we present this solution developed as a tool and show some preliminary
experiment results with real transaction datasets.
| no_new_dataset | 0.945901 |
1609.00992 | Nhien-An Le-Khac | Maarten Banerveld, Nhien-An Le-Khac, Tahar Kechadi | Performance Evaluation of a Natural Language Processing approach applied
in White Collar crime investigation | null | null | null | null | cs.IR cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In today world we are confronted with increasing amounts of information every
day coming from a large variety of sources. People and co-operations are
producing data on a large scale, and since the rise of the internet, e-mail and
social media the amount of produced data has grown exponentially. From a law
enforcement perspective we have to deal with these huge amounts of data when a
criminal investigation is launched against an individual or company. Relevant
questions need to be answered like who committed the crime, who were involved,
what happened and on what time, who were communicating and about what? Not only
the amount of available data to investigate has increased enormously, but also
the complexity of this data has increased. When these communication patterns
need to be combined with for instance a seized financial administration or
corporate document shares a complex investigation problem arises. Recently,
criminal investigators face a huge challenge when evidence of a crime needs to
be found in the Big Data environment where they have to deal with large and
complex datasets especially in financial and fraud investigations. To tackle
this problem, a financial and fraud investigation unit of a European country
has developed a new tool named LES that uses Natural Language Processing (NLP)
techniques to help criminal investigators handle large amounts of textual
information in a more efficient and faster way. In this paper, we present
briefly this tool and we focus on the evaluation its performance in terms of
the requirements of forensic investigation: speed, smarter and easier for
investigators. In order to evaluate this LES tool, we use different performance
metrics. We also show experimental results of our evaluation with large and
complex datasets from real-world application.
| [
{
"version": "v1",
"created": "Sun, 4 Sep 2016 21:23:22 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Banerveld",
"Maarten",
""
],
[
"Le-Khac",
"Nhien-An",
""
],
[
"Kechadi",
"Tahar",
""
]
] | TITLE: Performance Evaluation of a Natural Language Processing approach applied
in White Collar crime investigation
ABSTRACT: In today world we are confronted with increasing amounts of information every
day coming from a large variety of sources. People and co-operations are
producing data on a large scale, and since the rise of the internet, e-mail and
social media the amount of produced data has grown exponentially. From a law
enforcement perspective we have to deal with these huge amounts of data when a
criminal investigation is launched against an individual or company. Relevant
questions need to be answered like who committed the crime, who were involved,
what happened and on what time, who were communicating and about what? Not only
the amount of available data to investigate has increased enormously, but also
the complexity of this data has increased. When these communication patterns
need to be combined with for instance a seized financial administration or
corporate document shares a complex investigation problem arises. Recently,
criminal investigators face a huge challenge when evidence of a crime needs to
be found in the Big Data environment where they have to deal with large and
complex datasets especially in financial and fraud investigations. To tackle
this problem, a financial and fraud investigation unit of a European country
has developed a new tool named LES that uses Natural Language Processing (NLP)
techniques to help criminal investigators handle large amounts of textual
information in a more efficient and faster way. In this paper, we present
briefly this tool and we focus on the evaluation its performance in terms of
the requirements of forensic investigation: speed, smarter and easier for
investigators. In order to evaluate this LES tool, we use different performance
metrics. We also show experimental results of our evaluation with large and
complex datasets from real-world application.
| no_new_dataset | 0.934395 |
1609.01228 | Michel Fornaciali | Afonso Menegola, Michel Fornaciali, Ramon Pires, Sandra Avila, Eduardo
Valle | Towards Automated Melanoma Screening: Exploring Transfer Learning
Schemes | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep learning is the current bet for image classification. Its greed for huge
amounts of annotated data limits its usage in medical imaging context. In this
scenario transfer learning appears as a prominent solution. In this report we
aim to clarify how transfer learning schemes may influence classification
results. We are particularly focused in the automated melanoma screening
problem, a case of medical imaging in which transfer learning is still not
widely used. We explored transfer with and without fine-tuning, sequential
transfers and usage of pre-trained models in general and specific datasets.
Although some issues remain open, our findings may drive future researches.
| [
{
"version": "v1",
"created": "Mon, 5 Sep 2016 17:31:15 GMT"
}
] | 2016-09-06T00:00:00 | [
[
"Menegola",
"Afonso",
""
],
[
"Fornaciali",
"Michel",
""
],
[
"Pires",
"Ramon",
""
],
[
"Avila",
"Sandra",
""
],
[
"Valle",
"Eduardo",
""
]
] | TITLE: Towards Automated Melanoma Screening: Exploring Transfer Learning
Schemes
ABSTRACT: Deep learning is the current bet for image classification. Its greed for huge
amounts of annotated data limits its usage in medical imaging context. In this
scenario transfer learning appears as a prominent solution. In this report we
aim to clarify how transfer learning schemes may influence classification
results. We are particularly focused in the automated melanoma screening
problem, a case of medical imaging in which transfer learning is still not
widely used. We explored transfer with and without fine-tuning, sequential
transfers and usage of pre-trained models in general and specific datasets.
Although some issues remain open, our findings may drive future researches.
| no_new_dataset | 0.942454 |
1604.06582 | Jacopo Cavazza | Jacopo Cavazza, Andrea Zunino, Marco San Biagio and Vittorio Murino | Kernelized Covariance for Action Recognition | Accepted paper at the 23rd International Conference on Pattern
Recognition (ICPR), Cancun, Mexico, 2016 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we aim at increasing the descriptive power of the covariance
matrix, limited in capturing linear mutual dependencies between variables only.
We present a rigorous and principled mathematical pipeline to recover the
kernel trick for computing the covariance matrix, enhancing it to model more
complex, non-linear relationships conveyed by the raw data. To this end, we
propose Kernelized-COV, which generalizes the original covariance
representation without compromising the efficiency of the computation. In the
experiments, we validate the proposed framework against many previous
approaches in the literature, scoring on par or superior with respect to the
state of the art on benchmark datasets for 3D action recognition.
| [
{
"version": "v1",
"created": "Fri, 22 Apr 2016 09:16:22 GMT"
},
{
"version": "v2",
"created": "Fri, 2 Sep 2016 12:05:09 GMT"
}
] | 2016-09-05T00:00:00 | [
[
"Cavazza",
"Jacopo",
""
],
[
"Zunino",
"Andrea",
""
],
[
"Biagio",
"Marco San",
""
],
[
"Murino",
"Vittorio",
""
]
] | TITLE: Kernelized Covariance for Action Recognition
ABSTRACT: In this paper we aim at increasing the descriptive power of the covariance
matrix, limited in capturing linear mutual dependencies between variables only.
We present a rigorous and principled mathematical pipeline to recover the
kernel trick for computing the covariance matrix, enhancing it to model more
complex, non-linear relationships conveyed by the raw data. To this end, we
propose Kernelized-COV, which generalizes the original covariance
representation without compromising the efficiency of the computation. In the
experiments, we validate the proposed framework against many previous
approaches in the literature, scoring on par or superior with respect to the
state of the art on benchmark datasets for 3D action recognition.
| no_new_dataset | 0.946941 |
1608.05971 | Mohsen Fayyaz | Mohsen Fayyaz, Mohammad Hajizadeh Saffar, Mohammad Sabokrou, Mahmood
Fathy, Reinhard Klette, Fay Huang | STFCN: Spatio-Temporal FCN for Semantic Video Segmentation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | This paper presents a novel method to involve both spatial and temporal
features for semantic video segmentation. Current work on convolutional neural
networks(CNNs) has shown that CNNs provide advanced spatial features supporting
a very good performance of solutions for both image and video analysis,
especially for the semantic segmentation task. We investigate how involving
temporal features also has a good effect on segmenting video data. We propose a
module based on a long short-term memory (LSTM) architecture of a recurrent
neural network for interpreting the temporal characteristics of video frames
over time. Our system takes as input frames of a video and produces a
correspondingly-sized output; for segmenting the video our method combines the
use of three components: First, the regional spatial features of frames are
extracted using a CNN; then, using LSTM the temporal features are added;
finally, by deconvolving the spatio-temporal features we produce pixel-wise
predictions. Our key insight is to build spatio-temporal convolutional networks
(spatio-temporal CNNs) that have an end-to-end architecture for semantic video
segmentation. We adapted fully some known convolutional network architectures
(such as FCN-AlexNet and FCN-VGG16), and dilated convolution into our
spatio-temporal CNNs. Our spatio-temporal CNNs achieve state-of-the-art
semantic segmentation, as demonstrated for the Camvid and NYUDv2 datasets.
| [
{
"version": "v1",
"created": "Sun, 21 Aug 2016 17:34:08 GMT"
},
{
"version": "v2",
"created": "Fri, 2 Sep 2016 15:51:49 GMT"
}
] | 2016-09-05T00:00:00 | [
[
"Fayyaz",
"Mohsen",
""
],
[
"Saffar",
"Mohammad Hajizadeh",
""
],
[
"Sabokrou",
"Mohammad",
""
],
[
"Fathy",
"Mahmood",
""
],
[
"Klette",
"Reinhard",
""
],
[
"Huang",
"Fay",
""
]
] | TITLE: STFCN: Spatio-Temporal FCN for Semantic Video Segmentation
ABSTRACT: This paper presents a novel method to involve both spatial and temporal
features for semantic video segmentation. Current work on convolutional neural
networks(CNNs) has shown that CNNs provide advanced spatial features supporting
a very good performance of solutions for both image and video analysis,
especially for the semantic segmentation task. We investigate how involving
temporal features also has a good effect on segmenting video data. We propose a
module based on a long short-term memory (LSTM) architecture of a recurrent
neural network for interpreting the temporal characteristics of video frames
over time. Our system takes as input frames of a video and produces a
correspondingly-sized output; for segmenting the video our method combines the
use of three components: First, the regional spatial features of frames are
extracted using a CNN; then, using LSTM the temporal features are added;
finally, by deconvolving the spatio-temporal features we produce pixel-wise
predictions. Our key insight is to build spatio-temporal convolutional networks
(spatio-temporal CNNs) that have an end-to-end architecture for semantic video
segmentation. We adapted fully some known convolutional network architectures
(such as FCN-AlexNet and FCN-VGG16), and dilated convolution into our
spatio-temporal CNNs. Our spatio-temporal CNNs achieve state-of-the-art
semantic segmentation, as demonstrated for the Camvid and NYUDv2 datasets.
| no_new_dataset | 0.951006 |
1609.00426 | Paul M. Aoki | Paul M. Aoki | New Rain Rate Statistics for Emerging Regions: Implications for Wireless
Backhaul Planning | 14 pages; 3 figures | null | null | null | cs.NI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | As demand for broadband service increases in emerging regions, high-capacity
wireless links can accelerate and cost-reduce the deployment of new networks
(both backhaul and customer site connection). Such links are increasingly
common in developed countries, but their reliability in emerging regions is
questioned where very heavy tropical rain is present. Here, we investigate the
robustness of the standard (ITU-R P.837-6) method for estimating rain rates
using an expanded test dataset. We illustrate how bias/variance issues cause
problematic predictions at higher rain rates. We confirm (by construction) that
an improved rainfall climatology can largely address these prediction issues
without compromising standard ITU fit evaluation metrics.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 23:37:04 GMT"
}
] | 2016-09-05T00:00:00 | [
[
"Aoki",
"Paul M.",
""
]
] | TITLE: New Rain Rate Statistics for Emerging Regions: Implications for Wireless
Backhaul Planning
ABSTRACT: As demand for broadband service increases in emerging regions, high-capacity
wireless links can accelerate and cost-reduce the deployment of new networks
(both backhaul and customer site connection). Such links are increasingly
common in developed countries, but their reliability in emerging regions is
questioned where very heavy tropical rain is present. Here, we investigate the
robustness of the standard (ITU-R P.837-6) method for estimating rain rates
using an expanded test dataset. We illustrate how bias/variance issues cause
problematic predictions at higher rain rates. We confirm (by construction) that
an improved rainfall climatology can largely address these prediction issues
without compromising standard ITU fit evaluation metrics.
| new_dataset | 0.604107 |
1609.00462 | Markus Wagner | Markus Wagner, Marius Lindauer, Mustafa Misir, Samadhi Nallaperuma,
Frank Hutter | A case study of algorithm selection for the traveling thief problem | 23 pages, this article is underview | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Many real-world problems are composed of several interacting components. In
order to facilitate research on such interactions, the Traveling Thief Problem
(TTP) was created in 2013 as the combination of two well-understood
combinatorial optimization problems.
With this article, we contribute in four ways. First, we create a
comprehensive dataset that comprises the performance data of 21 TTP algorithms
on the full original set of 9720 TTP instances. Second, we define 55
characteristics for all TPP instances that can be used to select the best
algorithm on a per-instance basis. Third, we use these algorithms and features
to construct the first algorithm portfolios for TTP, clearly outperforming the
single best algorithm. Finally, we study which algorithms contribute most to
this portfolio.
| [
{
"version": "v1",
"created": "Fri, 2 Sep 2016 04:03:22 GMT"
}
] | 2016-09-05T00:00:00 | [
[
"Wagner",
"Markus",
""
],
[
"Lindauer",
"Marius",
""
],
[
"Misir",
"Mustafa",
""
],
[
"Nallaperuma",
"Samadhi",
""
],
[
"Hutter",
"Frank",
""
]
] | TITLE: A case study of algorithm selection for the traveling thief problem
ABSTRACT: Many real-world problems are composed of several interacting components. In
order to facilitate research on such interactions, the Traveling Thief Problem
(TTP) was created in 2013 as the combination of two well-understood
combinatorial optimization problems.
With this article, we contribute in four ways. First, we create a
comprehensive dataset that comprises the performance data of 21 TTP algorithms
on the full original set of 9720 TTP instances. Second, we define 55
characteristics for all TPP instances that can be used to select the best
algorithm on a per-instance basis. Third, we use these algorithms and features
to construct the first algorithm portfolios for TTP, clearly outperforming the
single best algorithm. Finally, we study which algorithms contribute most to
this portfolio.
| new_dataset | 0.951278 |
1609.00683 | Alessandro Checco | Alessandro Checco, Gianluca Demartini | Pairwise, Magnitude, or Stars: What's the Best Way for Crowds to Rate? | null | null | null | null | cs.IR cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We compare three popular techniques of rating content: the ubiquitous five
star rating, the less used pairwise comparison, and the recently introduced (in
crowdsourcing) magnitude estimation approach. Each system has specific
advantages and disadvantages, in terms of required user effort, achievable user
preference prediction accuracy and number of ratings required.
We design an experiment where the three techniques are compared in an
unbiased way. We collected 39'000 ratings on a popular crowdsourcing platform,
allowing us to release a dataset that will be useful for many related studies
on user rating techniques.
| [
{
"version": "v1",
"created": "Fri, 2 Sep 2016 17:50:53 GMT"
}
] | 2016-09-05T00:00:00 | [
[
"Checco",
"Alessandro",
""
],
[
"Demartini",
"Gianluca",
""
]
] | TITLE: Pairwise, Magnitude, or Stars: What's the Best Way for Crowds to Rate?
ABSTRACT: We compare three popular techniques of rating content: the ubiquitous five
star rating, the less used pairwise comparison, and the recently introduced (in
crowdsourcing) magnitude estimation approach. Each system has specific
advantages and disadvantages, in terms of required user effort, achievable user
preference prediction accuracy and number of ratings required.
We design an experiment where the three techniques are compared in an
unbiased way. We collected 39'000 ratings on a popular crowdsourcing platform,
allowing us to release a dataset that will be useful for many related studies
on user rating techniques.
| new_dataset | 0.941654 |
1609.00718 | Rie Johnson | Rie Johnson and Tong Zhang | Convolutional Neural Networks for Text Categorization: Shallow
Word-level vs. Deep Character-level | null | null | null | null | cs.CL cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper reports the performances of shallow word-level convolutional
neural networks (CNN), our earlier work (2015), on the eight datasets with
relatively large training data that were used for testing the very deep
character-level CNN in Conneau et al. (2016). Our findings are as follows. The
shallow word-level CNNs achieve better error rates than the error rates
reported in Conneau et al., though the results should be interpreted with some
consideration due to the unique pre-processing of Conneau et al. The shallow
word-level CNN uses more parameters and therefore requires more storage than
the deep character-level CNN; however, the shallow word-level CNN computes much
faster.
| [
{
"version": "v1",
"created": "Wed, 31 Aug 2016 15:43:27 GMT"
}
] | 2016-09-05T00:00:00 | [
[
"Johnson",
"Rie",
""
],
[
"Zhang",
"Tong",
""
]
] | TITLE: Convolutional Neural Networks for Text Categorization: Shallow
Word-level vs. Deep Character-level
ABSTRACT: This paper reports the performances of shallow word-level convolutional
neural networks (CNN), our earlier work (2015), on the eight datasets with
relatively large training data that were used for testing the very deep
character-level CNN in Conneau et al. (2016). Our findings are as follows. The
shallow word-level CNNs achieve better error rates than the error rates
reported in Conneau et al., though the results should be interpreted with some
consideration due to the unique pre-processing of Conneau et al. The shallow
word-level CNN uses more parameters and therefore requires more storage than
the deep character-level CNN; however, the shallow word-level CNN computes much
faster.
| no_new_dataset | 0.956796 |
1608.08337 | Avleen Bijral | Avleen S. Bijral | Data Dependent Convergence for Distributed Stochastic Optimization | PhD thesis | null | null | null | math.OC cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this dissertation we propose alternative analysis of distributed
stochastic gradient descent (SGD) algorithms that rely on spectral properties
of the data covariance. As a consequence we can relate questions pertaining to
speedups and convergence rates for distributed SGD to the data distribution
instead of the regularity properties of the objective functions. More precisely
we show that this rate depends on the spectral norm of the sample covariance
matrix. An estimate of this norm can provide practitioners with guidance
towards a potential gain in algorithm performance. For example many sparse
datasets with low spectral norm prove to be amenable to gains in distributed
settings. Towards establishing this data dependence we first study a
distributed consensus-based SGD algorithm and show that the rate of convergence
involves the spectral norm of the sample covariance matrix when the underlying
data is assumed to be independent and identically distributed (homogenous).
This dependence allows us to identify network regimes that prove to be
beneficial for datasets with low sample covariance spectral norm. Existing
consensus based analyses prove to be sub-optimal in the homogenous setting. Our
analysis method also allows us to find data-dependent convergence rates as we
limit the amount of communication. Spreading a fixed amount of data across more
nodes slows convergence; in the asymptotic regime we show that adding more
machines can help when minimizing twice-differentiable losses. Since the
mini-batch results don't follow from the consensus results we propose a
different data dependent analysis thereby providing theoretical validation for
why certain datasets are more amenable to mini-batching. We also provide
empirical evidence for results in this thesis.
| [
{
"version": "v1",
"created": "Tue, 30 Aug 2016 05:58:38 GMT"
}
] | 2016-09-03T00:00:00 | [
[
"Bijral",
"Avleen S.",
""
]
] | TITLE: Data Dependent Convergence for Distributed Stochastic Optimization
ABSTRACT: In this dissertation we propose alternative analysis of distributed
stochastic gradient descent (SGD) algorithms that rely on spectral properties
of the data covariance. As a consequence we can relate questions pertaining to
speedups and convergence rates for distributed SGD to the data distribution
instead of the regularity properties of the objective functions. More precisely
we show that this rate depends on the spectral norm of the sample covariance
matrix. An estimate of this norm can provide practitioners with guidance
towards a potential gain in algorithm performance. For example many sparse
datasets with low spectral norm prove to be amenable to gains in distributed
settings. Towards establishing this data dependence we first study a
distributed consensus-based SGD algorithm and show that the rate of convergence
involves the spectral norm of the sample covariance matrix when the underlying
data is assumed to be independent and identically distributed (homogenous).
This dependence allows us to identify network regimes that prove to be
beneficial for datasets with low sample covariance spectral norm. Existing
consensus based analyses prove to be sub-optimal in the homogenous setting. Our
analysis method also allows us to find data-dependent convergence rates as we
limit the amount of communication. Spreading a fixed amount of data across more
nodes slows convergence; in the asymptotic regime we show that adding more
machines can help when minimizing twice-differentiable losses. Since the
mini-batch results don't follow from the consensus results we propose a
different data dependent analysis thereby providing theoretical validation for
why certain datasets are more amenable to mini-batching. We also provide
empirical evidence for results in this thesis.
| no_new_dataset | 0.950778 |
1603.00275 | Korsuk Sirinukunwattana | Korsuk Sirinukunwattana, Josien P. W. Pluim, Hao Chen, Xiaojuan Qi,
Pheng-Ann Heng, Yun Bo Guo, Li Yang Wang, Bogdan J. Matuszewski, Elia Bruni,
Urko Sanchez, Anton B\"ohm, Olaf Ronneberger, Bassem Ben Cheikh, Daniel
Racoceanu, Philipp Kainz, Michael Pfeiffer, Martin Urschler, David R. J.
Snead, Nasir M. Rajpoot | Gland Segmentation in Colon Histology Images: The GlaS Challenge Contest | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Colorectal adenocarcinoma originating in intestinal glandular structures is
the most common form of colon cancer. In clinical practice, the morphology of
intestinal glands, including architectural appearance and glandular formation,
is used by pathologists to inform prognosis and plan the treatment of
individual patients. However, achieving good inter-observer as well as
intra-observer reproducibility of cancer grading is still a major challenge in
modern pathology. An automated approach which quantifies the morphology of
glands is a solution to the problem. This paper provides an overview to the
Gland Segmentation in Colon Histology Images Challenge Contest (GlaS) held at
MICCAI'2015. Details of the challenge, including organization, dataset and
evaluation criteria, are presented, along with the method descriptions and
evaluation results from the top performing methods.
| [
{
"version": "v1",
"created": "Tue, 1 Mar 2016 13:53:48 GMT"
},
{
"version": "v2",
"created": "Thu, 1 Sep 2016 14:18:51 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Sirinukunwattana",
"Korsuk",
""
],
[
"Pluim",
"Josien P. W.",
""
],
[
"Chen",
"Hao",
""
],
[
"Qi",
"Xiaojuan",
""
],
[
"Heng",
"Pheng-Ann",
""
],
[
"Guo",
"Yun Bo",
""
],
[
"Wang",
"Li Yang",
""
],
[
"Matuszewski",
"Bogdan J.",
""
],
[
"Bruni",
"Elia",
""
],
[
"Sanchez",
"Urko",
""
],
[
"Böhm",
"Anton",
""
],
[
"Ronneberger",
"Olaf",
""
],
[
"Cheikh",
"Bassem Ben",
""
],
[
"Racoceanu",
"Daniel",
""
],
[
"Kainz",
"Philipp",
""
],
[
"Pfeiffer",
"Michael",
""
],
[
"Urschler",
"Martin",
""
],
[
"Snead",
"David R. J.",
""
],
[
"Rajpoot",
"Nasir M.",
""
]
] | TITLE: Gland Segmentation in Colon Histology Images: The GlaS Challenge Contest
ABSTRACT: Colorectal adenocarcinoma originating in intestinal glandular structures is
the most common form of colon cancer. In clinical practice, the morphology of
intestinal glands, including architectural appearance and glandular formation,
is used by pathologists to inform prognosis and plan the treatment of
individual patients. However, achieving good inter-observer as well as
intra-observer reproducibility of cancer grading is still a major challenge in
modern pathology. An automated approach which quantifies the morphology of
glands is a solution to the problem. This paper provides an overview to the
Gland Segmentation in Colon Histology Images Challenge Contest (GlaS) held at
MICCAI'2015. Details of the challenge, including organization, dataset and
evaluation criteria, are presented, along with the method descriptions and
evaluation results from the top performing methods.
| no_new_dataset | 0.939748 |
1605.01379 | Xiao Lin | Xiao Lin, Devi Parikh | Leveraging Visual Question Answering for Image-Caption Ranking | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual Question Answering (VQA) is the task of taking as input an image and a
free-form natural language question about the image, and producing an accurate
answer. In this work we view VQA as a "feature extraction" module to extract
image and caption representations. We employ these representations for the task
of image-caption ranking. Each feature dimension captures (imagines) whether a
fact (question-answer pair) could plausibly be true for the image and caption.
This allows the model to interpret images and captions from a wide variety of
perspectives. We propose score-level and representation-level fusion models to
incorporate VQA knowledge in an existing state-of-the-art VQA-agnostic
image-caption ranking model. We find that incorporating and reasoning about
consistency between images and captions significantly improves performance.
Concretely, our model improves state-of-the-art on caption retrieval by 7.1%
and on image retrieval by 4.4% on the MSCOCO dataset.
| [
{
"version": "v1",
"created": "Wed, 4 May 2016 18:54:09 GMT"
},
{
"version": "v2",
"created": "Wed, 31 Aug 2016 20:14:12 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Lin",
"Xiao",
""
],
[
"Parikh",
"Devi",
""
]
] | TITLE: Leveraging Visual Question Answering for Image-Caption Ranking
ABSTRACT: Visual Question Answering (VQA) is the task of taking as input an image and a
free-form natural language question about the image, and producing an accurate
answer. In this work we view VQA as a "feature extraction" module to extract
image and caption representations. We employ these representations for the task
of image-caption ranking. Each feature dimension captures (imagines) whether a
fact (question-answer pair) could plausibly be true for the image and caption.
This allows the model to interpret images and captions from a wide variety of
perspectives. We propose score-level and representation-level fusion models to
incorporate VQA knowledge in an existing state-of-the-art VQA-agnostic
image-caption ranking model. We find that incorporating and reasoning about
consistency between images and captions significantly improves performance.
Concretely, our model improves state-of-the-art on caption retrieval by 7.1%
and on image retrieval by 4.4% on the MSCOCO dataset.
| no_new_dataset | 0.948394 |
1607.06275 | Peng Li | Peng Li, Wei Li, Zhengyan He, Xuguang Wang, Ying Cao, Jie Zhou, Wei Xu | Dataset and Neural Recurrent Sequence Labeling Model for Open-Domain
Factoid Question Answering | 10 pages, 3 figures, withdraw experimental results on CNN/Daily Mail
datasets | null | null | null | cs.CL cs.AI cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While question answering (QA) with neural network, i.e. neural QA, has
achieved promising results in recent years, lacking of large scale real-word QA
dataset is still a challenge for developing and evaluating neural QA system. To
alleviate this problem, we propose a large scale human annotated real-world QA
dataset WebQA with more than 42k questions and 556k evidences. As existing
neural QA methods resolve QA either as sequence generation or
classification/ranking problem, they face challenges of expensive softmax
computation, unseen answers handling or separate candidate answer generation
component. In this work, we cast neural QA as a sequence labeling problem and
propose an end-to-end sequence labeling model, which overcomes all the above
challenges. Experimental results on WebQA show that our model outperforms the
baselines significantly with an F1 score of 74.69% with word-based input, and
the performance drops only 3.72 F1 points with more challenging character-based
input.
| [
{
"version": "v1",
"created": "Thu, 21 Jul 2016 11:40:50 GMT"
},
{
"version": "v2",
"created": "Thu, 1 Sep 2016 10:56:45 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Li",
"Peng",
""
],
[
"Li",
"Wei",
""
],
[
"He",
"Zhengyan",
""
],
[
"Wang",
"Xuguang",
""
],
[
"Cao",
"Ying",
""
],
[
"Zhou",
"Jie",
""
],
[
"Xu",
"Wei",
""
]
] | TITLE: Dataset and Neural Recurrent Sequence Labeling Model for Open-Domain
Factoid Question Answering
ABSTRACT: While question answering (QA) with neural network, i.e. neural QA, has
achieved promising results in recent years, lacking of large scale real-word QA
dataset is still a challenge for developing and evaluating neural QA system. To
alleviate this problem, we propose a large scale human annotated real-world QA
dataset WebQA with more than 42k questions and 556k evidences. As existing
neural QA methods resolve QA either as sequence generation or
classification/ranking problem, they face challenges of expensive softmax
computation, unseen answers handling or separate candidate answer generation
component. In this work, we cast neural QA as a sequence labeling problem and
propose an end-to-end sequence labeling model, which overcomes all the above
challenges. Experimental results on WebQA show that our model outperforms the
baselines significantly with an F1 score of 74.69% with word-based input, and
the performance drops only 3.72 F1 points with more challenging character-based
input.
| new_dataset | 0.750827 |
1608.07793 | Zhongqi Lu | Zhongqi Lu, Qiang Yang | Partially Observable Markov Decision Process for Recommender Systems | null | null | null | null | cs.AI cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We report the "Recurrent Deterioration" (RD) phenomenon observed in online
recommender systems. The RD phenomenon is reflected by the trend of performance
degradation when the recommendation model is always trained based on users'
feedbacks of the previous recommendations. There are several reasons for the
recommender systems to encounter the RD phenomenon, including the lack of
negative training data and the evolution of users' interests, etc. Motivated to
tackle the problems causing the RD phenomenon, we propose the POMDP-Rec
framework, which is a neural-optimized Partially Observable Markov Decision
Process algorithm for recommender systems. We show that the POMDP-Rec framework
effectively uses the accumulated historical data from real-world recommender
systems and automatically achieves comparable results with those models
fine-tuned exhaustively by domain exports on public datasets.
| [
{
"version": "v1",
"created": "Sun, 28 Aug 2016 09:42:52 GMT"
},
{
"version": "v2",
"created": "Thu, 1 Sep 2016 15:41:02 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Lu",
"Zhongqi",
""
],
[
"Yang",
"Qiang",
""
]
] | TITLE: Partially Observable Markov Decision Process for Recommender Systems
ABSTRACT: We report the "Recurrent Deterioration" (RD) phenomenon observed in online
recommender systems. The RD phenomenon is reflected by the trend of performance
degradation when the recommendation model is always trained based on users'
feedbacks of the previous recommendations. There are several reasons for the
recommender systems to encounter the RD phenomenon, including the lack of
negative training data and the evolution of users' interests, etc. Motivated to
tackle the problems causing the RD phenomenon, we propose the POMDP-Rec
framework, which is a neural-optimized Partially Observable Markov Decision
Process algorithm for recommender systems. We show that the POMDP-Rec framework
effectively uses the accumulated historical data from real-world recommender
systems and automatically achieves comparable results with those models
fine-tuned exhaustively by domain exports on public datasets.
| no_new_dataset | 0.949623 |
1608.08967 | Seyed-Mohsen Moosavi-Dezfooli | Alhussein Fawzi, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard | Robustness of classifiers: from adversarial to random noise | Accepted to NIPS 2016 | null | null | null | cs.LG cs.CV stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Several recent works have shown that state-of-the-art classifiers are
vulnerable to worst-case (i.e., adversarial) perturbations of the datapoints.
On the other hand, it has been empirically observed that these same classifiers
are relatively robust to random noise. In this paper, we propose to study a
\textit{semi-random} noise regime that generalizes both the random and
worst-case noise regimes. We propose the first quantitative analysis of the
robustness of nonlinear classifiers in this general noise regime. We establish
precise theoretical bounds on the robustness of classifiers in this general
regime, which depend on the curvature of the classifier's decision boundary.
Our bounds confirm and quantify the empirical observations that classifiers
satisfying curvature constraints are robust to random noise. Moreover, we
quantify the robustness of classifiers in terms of the subspace dimension in
the semi-random noise regime, and show that our bounds remarkably interpolate
between the worst-case and random noise regimes. We perform experiments and
show that the derived bounds provide very accurate estimates when applied to
various state-of-the-art deep neural networks and datasets. This result
suggests bounds on the curvature of the classifiers' decision boundaries that
we support experimentally, and more generally offers important insights onto
the geometry of high dimensional classification problems.
| [
{
"version": "v1",
"created": "Wed, 31 Aug 2016 17:54:34 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Fawzi",
"Alhussein",
""
],
[
"Moosavi-Dezfooli",
"Seyed-Mohsen",
""
],
[
"Frossard",
"Pascal",
""
]
] | TITLE: Robustness of classifiers: from adversarial to random noise
ABSTRACT: Several recent works have shown that state-of-the-art classifiers are
vulnerable to worst-case (i.e., adversarial) perturbations of the datapoints.
On the other hand, it has been empirically observed that these same classifiers
are relatively robust to random noise. In this paper, we propose to study a
\textit{semi-random} noise regime that generalizes both the random and
worst-case noise regimes. We propose the first quantitative analysis of the
robustness of nonlinear classifiers in this general noise regime. We establish
precise theoretical bounds on the robustness of classifiers in this general
regime, which depend on the curvature of the classifier's decision boundary.
Our bounds confirm and quantify the empirical observations that classifiers
satisfying curvature constraints are robust to random noise. Moreover, we
quantify the robustness of classifiers in terms of the subspace dimension in
the semi-random noise regime, and show that our bounds remarkably interpolate
between the worst-case and random noise regimes. We perform experiments and
show that the derived bounds provide very accurate estimates when applied to
various state-of-the-art deep neural networks and datasets. This result
suggests bounds on the curvature of the classifiers' decision boundaries that
we support experimentally, and more generally offers important insights onto
the geometry of high dimensional classification problems.
| no_new_dataset | 0.950134 |
1609.00017 | Gordon Christie | Gordon Christie, Adam Shoemaker, Kevin Kochersberger, Pratap Tokekar,
Lance McLean, Alexander Leonessa | Radiation Search Operations using Scene Understanding with Autonomous
UAV and UGV | null | null | null | null | cs.RO cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Autonomously searching for hazardous radiation sources requires the ability
of the aerial and ground systems to understand the scene they are scouting. In
this paper, we present systems, algorithms, and experiments to perform
radiation search using unmanned aerial vehicles (UAV) and unmanned ground
vehicles (UGV) by employing semantic scene segmentation. The aerial data is
used to identify radiological points of interest, generate an orthophoto along
with a digital elevation model (DEM) of the scene, and perform semantic
segmentation to assign a category (e.g. road, grass) to each pixel in the
orthophoto. We perform semantic segmentation by training a model on a dataset
of images we collected and annotated, using the model to perform inference on
images of the test area unseen to the model, and then refining the results with
the DEM to better reason about category predictions at each pixel. We then use
all of these outputs to plan a path for a UGV carrying a LiDAR to map the
environment and avoid obstacles not present during the flight, and a radiation
detector to collect more precise radiation measurements from the ground.
Results of the analysis for each scenario tested favorably. We also note that
our approach is general and has the potential to work for a variety of
different sensing tasks.
| [
{
"version": "v1",
"created": "Wed, 31 Aug 2016 20:00:46 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Christie",
"Gordon",
""
],
[
"Shoemaker",
"Adam",
""
],
[
"Kochersberger",
"Kevin",
""
],
[
"Tokekar",
"Pratap",
""
],
[
"McLean",
"Lance",
""
],
[
"Leonessa",
"Alexander",
""
]
] | TITLE: Radiation Search Operations using Scene Understanding with Autonomous
UAV and UGV
ABSTRACT: Autonomously searching for hazardous radiation sources requires the ability
of the aerial and ground systems to understand the scene they are scouting. In
this paper, we present systems, algorithms, and experiments to perform
radiation search using unmanned aerial vehicles (UAV) and unmanned ground
vehicles (UGV) by employing semantic scene segmentation. The aerial data is
used to identify radiological points of interest, generate an orthophoto along
with a digital elevation model (DEM) of the scene, and perform semantic
segmentation to assign a category (e.g. road, grass) to each pixel in the
orthophoto. We perform semantic segmentation by training a model on a dataset
of images we collected and annotated, using the model to perform inference on
images of the test area unseen to the model, and then refining the results with
the DEM to better reason about category predictions at each pixel. We then use
all of these outputs to plan a path for a UGV carrying a LiDAR to map the
environment and avoid obstacles not present during the flight, and a radiation
detector to collect more precise radiation measurements from the ground.
Results of the analysis for each scenario tested favorably. We also note that
our approach is general and has the potential to work for a variety of
different sensing tasks.
| no_new_dataset | 0.815673 |
1609.00070 | Arun Tejasvi Chaganty | Arun Tejasvi Chaganty and Percy Liang | How Much is 131 Million Dollars? Putting Numbers in Perspective with
Compositional Descriptions | null | ACL (2016), 578-587 | 10.18653/v1/P16-1055 | null | cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | How much is 131 million US dollars? To help readers put such numbers in
context, we propose a new task of automatically generating short descriptions
known as perspectives, e.g. "$131 million is about the cost to employ everyone
in Texas over a lunch period". First, we collect a dataset of numeric mentions
in news articles, where each mention is labeled with a set of rated
perspectives. We then propose a system to generate these descriptions
consisting of two steps: formula construction and description generation. In
construction, we compose formulae from numeric facts in a knowledge base and
rank the resulting formulas based on familiarity, numeric proximity and
semantic compatibility. In generation, we convert a formula into natural
language using a sequence-to-sequence recurrent neural network. Our system
obtains a 15.2% F1 improvement over a non-compositional baseline at formula
construction and a 12.5 BLEU point improvement over a baseline description
generation.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 00:20:41 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Chaganty",
"Arun Tejasvi",
""
],
[
"Liang",
"Percy",
""
]
] | TITLE: How Much is 131 Million Dollars? Putting Numbers in Perspective with
Compositional Descriptions
ABSTRACT: How much is 131 million US dollars? To help readers put such numbers in
context, we propose a new task of automatically generating short descriptions
known as perspectives, e.g. "$131 million is about the cost to employ everyone
in Texas over a lunch period". First, we collect a dataset of numeric mentions
in news articles, where each mention is labeled with a set of rated
perspectives. We then propose a system to generate these descriptions
consisting of two steps: formula construction and description generation. In
construction, we compose formulae from numeric facts in a knowledge base and
rank the resulting formulas based on familiarity, numeric proximity and
semantic compatibility. In generation, we convert a formula into natural
language using a sequence-to-sequence recurrent neural network. Our system
obtains a 15.2% F1 improvement over a non-compositional baseline at formula
construction and a 12.5 BLEU point improvement over a baseline description
generation.
| no_new_dataset | 0.733452 |
1609.00081 | Tanmoy Chakraborty | Tanmoy Chakraborty and Ramasuri Narayanam | All Fingers are not Equal: Intensity of References in Scientific
Articles | 11 pages, 4 figures, 4 tables, Conference on Empirical Methods in
Natural Language Processing (EMNLP 2016) | null | null | null | cs.CL cs.DL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Research accomplishment is usually measured by considering all citations with
equal importance, thus ignoring the wide variety of purposes an article is
being cited for. Here, we posit that measuring the intensity of a reference is
crucial not only to perceive better understanding of research endeavor, but
also to improve the quality of citation-based applications. To this end, we
collect a rich annotated dataset with references labeled by the intensity, and
propose a novel graph-based semi-supervised model, GraLap to label the
intensity of references. Experiments with AAN datasets show a significant
improvement compared to the baselines to achieve the true labels of the
references (46% better correlation). Finally, we provide four applications to
demonstrate how the knowledge of reference intensity leads to design better
real-world applications.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 01:34:56 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Chakraborty",
"Tanmoy",
""
],
[
"Narayanam",
"Ramasuri",
""
]
] | TITLE: All Fingers are not Equal: Intensity of References in Scientific
Articles
ABSTRACT: Research accomplishment is usually measured by considering all citations with
equal importance, thus ignoring the wide variety of purposes an article is
being cited for. Here, we posit that measuring the intensity of a reference is
crucial not only to perceive better understanding of research endeavor, but
also to improve the quality of citation-based applications. To this end, we
collect a rich annotated dataset with references labeled by the intensity, and
propose a novel graph-based semi-supervised model, GraLap to label the
intensity of references. Experiments with AAN datasets show a significant
improvement compared to the baselines to achieve the true labels of the
references (46% better correlation). Finally, we provide four applications to
demonstrate how the knowledge of reference intensity leads to design better
real-world applications.
| new_dataset | 0.962108 |
1609.00130 | Roberto Rigamonti | Roberto Rigamonti and Baptiste Delporte and Anthony Convers and
Alberto Dassatti | Transparent Live Code Offloading on FPGA | 9 pages in FPGAs for Software Programmers (FSP 2016) | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Even though it seems that FPGAs have finally made the transition from
research labs to the consumer devices' market, programming them remains
challenging. Despite the improvements made by High-Level Synthesis (HLS), which
removed the language and paradigm barriers that prevented many computer
scientists from working with them, producing a new design typically requires at
least several hours, making data- and context-dependent adaptations virtually
impossible.
In this paper we present a new framework that off-loads, on-the-fly and
transparently to both the user and the developer, computationally-intensive
code fragments to FPGAs. While the performance should not surpass that of
hand-crafted HDL code, or even code produced by HLS, our results come with no
additional development costs and do not require producing and deploying a new
bit-stream to the FPGA each time a change is made. Moreover, since
optimizations are made at run-time, they may fit particular datasets or usage
scenarios, something which is rarely foreseeable at design or compile time.
Our proposal revolves around an overlay architecture that is pre-programmed
on the FPGA and dynamically reconfigured by our framework to execute code
fragments extracted from the Data Flow Graph (DFG) of computational intensive
routines. We validated our solution using standard benchmarks and proved we are
able to off-load to FPGAs without developer's intervention.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 07:18:41 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Rigamonti",
"Roberto",
""
],
[
"Delporte",
"Baptiste",
""
],
[
"Convers",
"Anthony",
""
],
[
"Dassatti",
"Alberto",
""
]
] | TITLE: Transparent Live Code Offloading on FPGA
ABSTRACT: Even though it seems that FPGAs have finally made the transition from
research labs to the consumer devices' market, programming them remains
challenging. Despite the improvements made by High-Level Synthesis (HLS), which
removed the language and paradigm barriers that prevented many computer
scientists from working with them, producing a new design typically requires at
least several hours, making data- and context-dependent adaptations virtually
impossible.
In this paper we present a new framework that off-loads, on-the-fly and
transparently to both the user and the developer, computationally-intensive
code fragments to FPGAs. While the performance should not surpass that of
hand-crafted HDL code, or even code produced by HLS, our results come with no
additional development costs and do not require producing and deploying a new
bit-stream to the FPGA each time a change is made. Moreover, since
optimizations are made at run-time, they may fit particular datasets or usage
scenarios, something which is rarely foreseeable at design or compile time.
Our proposal revolves around an overlay architecture that is pre-programmed
on the FPGA and dynamically reconfigured by our framework to execute code
fragments extracted from the Data Flow Graph (DFG) of computational intensive
routines. We validated our solution using standard benchmarks and proved we are
able to off-load to FPGAs without developer's intervention.
| no_new_dataset | 0.940463 |
1609.00154 | Sebastian Sippel | Sebastian Sippel and Jakob Zscheischler and Martin Heimann and Holger
Lange and Miguel D. Mahecha and Geert Jan van Oldenborgh and Friederike E.L.
Otto and Markus Reichstein | Have precipitation extremes and annual totals been increasing in the
world's dry regions over the last 60 years? | null | null | null | null | physics.geo-ph physics.ao-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Daily rainfall extremes and annual totals have increased in large parts of
the global land area over the last decades. These observations are consistent
with theoretical considerations of a warming climate. However, until recently
these global tendencies have not been shown to consistently affect land regions
with limited moisture availability. A recent study, published by Donat et al.
(2016, Nature Climate Change, doi:10.1038/nclimate2941), now identified rapid
increases in globally aggregated dry region daily extreme rainfall and annual
rainfall totals. Here, we reassess the respective analysis and find that a)
statistical artifacts introduced by the choice of the reference period prior to
data standardization lead to an overestimation of the reported trends by up to
40%, and also that b) the definition of `dry regions of the globe' affect the
reported globally aggregated trends in extreme rainfall. Using the same
observational dataset, but accounting for the statistical artifacts and using
alternative, well-established dryness definitions, we find no significant
increases in heavy precipitation in the world's dry regions. Adequate data
pre-processing approaches and accounting for uncertainties regarding the
definition of dryness are crucial to the quantification of spatially aggregated
trends in the world's dry regions. In view of the high relevance of the
question to many potentially affected stakeholders, we call for a cautionary
consideration of specific data processing methods, including issues related to
the definition of dry areas, to guarantee robustness of communicated climate
change relevant findings.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 09:20:20 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Sippel",
"Sebastian",
""
],
[
"Zscheischler",
"Jakob",
""
],
[
"Heimann",
"Martin",
""
],
[
"Lange",
"Holger",
""
],
[
"Mahecha",
"Miguel D.",
""
],
[
"van Oldenborgh",
"Geert Jan",
""
],
[
"Otto",
"Friederike E. L.",
""
],
[
"Reichstein",
"Markus",
""
]
] | TITLE: Have precipitation extremes and annual totals been increasing in the
world's dry regions over the last 60 years?
ABSTRACT: Daily rainfall extremes and annual totals have increased in large parts of
the global land area over the last decades. These observations are consistent
with theoretical considerations of a warming climate. However, until recently
these global tendencies have not been shown to consistently affect land regions
with limited moisture availability. A recent study, published by Donat et al.
(2016, Nature Climate Change, doi:10.1038/nclimate2941), now identified rapid
increases in globally aggregated dry region daily extreme rainfall and annual
rainfall totals. Here, we reassess the respective analysis and find that a)
statistical artifacts introduced by the choice of the reference period prior to
data standardization lead to an overestimation of the reported trends by up to
40%, and also that b) the definition of `dry regions of the globe' affect the
reported globally aggregated trends in extreme rainfall. Using the same
observational dataset, but accounting for the statistical artifacts and using
alternative, well-established dryness definitions, we find no significant
increases in heavy precipitation in the world's dry regions. Adequate data
pre-processing approaches and accounting for uncertainties regarding the
definition of dryness are crucial to the quantification of spatially aggregated
trends in the world's dry regions. In view of the high relevance of the
question to many potentially affected stakeholders, we call for a cautionary
consideration of specific data processing methods, including issues related to
the definition of dry areas, to guarantee robustness of communicated climate
change relevant findings.
| no_new_dataset | 0.941007 |
1609.00162 | Limin Wang | Limin Wang, Zhe Wang, Yu Qiao, Luc Van Gool | Transferring Object-Scene Convolutional Neural Networks for Event
Recognition in Still Images | 20 pages | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Event recognition in still images is an intriguing problem and has potential
for real applications. This paper addresses the problem of event recognition by
proposing a convolutional neural network that exploits knowledge of objects and
scenes for event classification (OS2E-CNN). Intuitively, it stands to reason
that there exists a correlation among the concepts of objects, scenes, and
events. We empirically demonstrate that the recognition of objects and scenes
substantially contributes to the recognition of events. Meanwhile, we propose
an iterative selection method to identify a subset of object and scene classes,
which help to more efficiently and effectively transfer their deep
representations to event recognition. Specifically, we develop three types of
transferring techniques: (1) initialization-based transferring, (2)
knowledge-based transferring, and (3) data-based transferring. These newly
designed transferring techniques exploit multi-task learning frameworks to
incorporate extra knowledge from other networks and additional datasets into
the training procedure of event CNNs. These multi-task learning frameworks turn
out to be effective in reducing the effect of over-fitting and improving the
generalization ability of the learned CNNs. With OS2E-CNN, we design a
multi-ratio and multi-scale cropping strategy, and propose an end-to-end event
recognition pipeline. We perform experiments on three event recognition
benchmarks: the ChaLearn Cultural Event Recognition dataset, the Web Image
Dataset for Event Recognition (WIDER), and the UIUC Sports Event dataset. The
experimental results show that our proposed algorithm successfully adapts
object and scene representations towards the event dataset and that it achieves
the current state-of-the-art performance on these challenging datasets.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 09:47:22 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Wang",
"Limin",
""
],
[
"Wang",
"Zhe",
""
],
[
"Qiao",
"Yu",
""
],
[
"Van Gool",
"Luc",
""
]
] | TITLE: Transferring Object-Scene Convolutional Neural Networks for Event
Recognition in Still Images
ABSTRACT: Event recognition in still images is an intriguing problem and has potential
for real applications. This paper addresses the problem of event recognition by
proposing a convolutional neural network that exploits knowledge of objects and
scenes for event classification (OS2E-CNN). Intuitively, it stands to reason
that there exists a correlation among the concepts of objects, scenes, and
events. We empirically demonstrate that the recognition of objects and scenes
substantially contributes to the recognition of events. Meanwhile, we propose
an iterative selection method to identify a subset of object and scene classes,
which help to more efficiently and effectively transfer their deep
representations to event recognition. Specifically, we develop three types of
transferring techniques: (1) initialization-based transferring, (2)
knowledge-based transferring, and (3) data-based transferring. These newly
designed transferring techniques exploit multi-task learning frameworks to
incorporate extra knowledge from other networks and additional datasets into
the training procedure of event CNNs. These multi-task learning frameworks turn
out to be effective in reducing the effect of over-fitting and improving the
generalization ability of the learned CNNs. With OS2E-CNN, we design a
multi-ratio and multi-scale cropping strategy, and propose an end-to-end event
recognition pipeline. We perform experiments on three event recognition
benchmarks: the ChaLearn Cultural Event Recognition dataset, the Web Image
Dataset for Event Recognition (WIDER), and the UIUC Sports Event dataset. The
experimental results show that our proposed algorithm successfully adapts
object and scene representations towards the event dataset and that it achieves
the current state-of-the-art performance on these challenging datasets.
| no_new_dataset | 0.948489 |
1609.00203 | Angelos Valsamis | Angelos Valsamis, Konstantinos Tserpes, Dimitrios Zissis, Dimosthenis
Anagnostopoulos, Theodora Varvarigou | Employing traditional machine learning algorithms for big data streams
analysis: the case of object trajectory prediction | 14 pages, 2 figures, 3 tables, 31 references | null | 10.1016/j.jss.2016.06.016 | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we model the trajectory of sea vessels and provide a service
that predicts in near-real time the position of any given vessel in 4', 10',
20' and 40' time intervals. We explore the necessary tradeoffs between
accuracy, performance and resource utilization are explored given the large
volume and update rates of input data. We start with building models based on
well-established machine learning algorithms using static datasets and
multi-scan training approaches and identify the best candidate to be used in
implementing a single-pass predictive approach, under real-time constraints.
The results are measured in terms of accuracy and performance and are compared
against the baseline kinematic equations. Results show that it is possible to
efficiently model the trajectory of multiple vessels using a single model,
which is trained and evaluated using an adequately large, static dataset, thus
achieving a significant gain in terms of resource usage while not compromising
accuracy.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 12:06:20 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Valsamis",
"Angelos",
""
],
[
"Tserpes",
"Konstantinos",
""
],
[
"Zissis",
"Dimitrios",
""
],
[
"Anagnostopoulos",
"Dimosthenis",
""
],
[
"Varvarigou",
"Theodora",
""
]
] | TITLE: Employing traditional machine learning algorithms for big data streams
analysis: the case of object trajectory prediction
ABSTRACT: In this paper, we model the trajectory of sea vessels and provide a service
that predicts in near-real time the position of any given vessel in 4', 10',
20' and 40' time intervals. We explore the necessary tradeoffs between
accuracy, performance and resource utilization are explored given the large
volume and update rates of input data. We start with building models based on
well-established machine learning algorithms using static datasets and
multi-scan training approaches and identify the best candidate to be used in
implementing a single-pass predictive approach, under real-time constraints.
The results are measured in terms of accuracy and performance and are compared
against the baseline kinematic equations. Results show that it is possible to
efficiently model the trajectory of multiple vessels using a single model,
which is trained and evaluated using an adequately large, static dataset, thus
achieving a significant gain in terms of resource usage while not compromising
accuracy.
| no_new_dataset | 0.943608 |
1609.00221 | Federico Becattini | Giovanni Cuffaro, Federico Becattini, Claudio Baecchi, Lorenzo
Seidenari, Alberto Del Bimbo | Segmentation Free Object Discovery in Video | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we present a simple yet effective approach to extend without
supervision any object proposal from static images to videos. Unlike previous
methods, these spatio-temporal proposals, to which we refer as tracks, are
generated relying on little or no visual content by only exploiting bounding
boxes spatial correlations through time. The tracks that we obtain are likely
to represent objects and are a general-purpose tool to represent meaningful
video content for a wide variety of tasks. For unannotated videos, tracks can
be used to discover content without any supervision. As further contribution we
also propose a novel and dataset-independent method to evaluate a generic
object proposal based on the entropy of a classifier output response. We
experiment on two competitive datasets, namely YouTube Objects and ILSVRC-2015
VID.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 13:08:39 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Cuffaro",
"Giovanni",
""
],
[
"Becattini",
"Federico",
""
],
[
"Baecchi",
"Claudio",
""
],
[
"Seidenari",
"Lorenzo",
""
],
[
"Del Bimbo",
"Alberto",
""
]
] | TITLE: Segmentation Free Object Discovery in Video
ABSTRACT: In this paper we present a simple yet effective approach to extend without
supervision any object proposal from static images to videos. Unlike previous
methods, these spatio-temporal proposals, to which we refer as tracks, are
generated relying on little or no visual content by only exploiting bounding
boxes spatial correlations through time. The tracks that we obtain are likely
to represent objects and are a general-purpose tool to represent meaningful
video content for a wide variety of tasks. For unannotated videos, tracks can
be used to discover content without any supervision. As further contribution we
also propose a novel and dataset-independent method to evaluate a generic
object proposal based on the entropy of a classifier output response. We
experiment on two competitive datasets, namely YouTube Objects and ILSVRC-2015
VID.
| no_new_dataset | 0.951369 |
1609.00278 | Arsalan Mousavian | Arsalan Mousavian, Jana Kosecka | Semantic Image Based Geolocation Given a Map | null | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The problem visual place recognition is commonly used strategy for
localization. Most successful appearance based methods typically rely on a
large database of views endowed with local or global image descriptors and
strive to retrieve the views of the same location. The quality of the results
is often affected by the density of the reference views and the robustness of
the image representation with respect to viewpoint variations, clutter and
seasonal changes. In this work we present an approach for geo-locating a novel
view and determining camera location and orientation using a map and a sparse
set of geo-tagged reference views. We propose a novel technique for detection
and identification of building facades from geo-tagged reference view using the
map and geometry of the building facades. We compute the likelihood of camera
location and orientation of the query images using the detected landmark
(building) identities from reference views, 2D map of the environment, and
geometry of building facades. We evaluate our approach for building
identification and geo-localization on a new challenging outdoors urban dataset
exhibiting large variations in appearance and viewpoint.
| [
{
"version": "v1",
"created": "Thu, 1 Sep 2016 15:27:02 GMT"
}
] | 2016-09-02T00:00:00 | [
[
"Mousavian",
"Arsalan",
""
],
[
"Kosecka",
"Jana",
""
]
] | TITLE: Semantic Image Based Geolocation Given a Map
ABSTRACT: The problem visual place recognition is commonly used strategy for
localization. Most successful appearance based methods typically rely on a
large database of views endowed with local or global image descriptors and
strive to retrieve the views of the same location. The quality of the results
is often affected by the density of the reference views and the robustness of
the image representation with respect to viewpoint variations, clutter and
seasonal changes. In this work we present an approach for geo-locating a novel
view and determining camera location and orientation using a map and a sparse
set of geo-tagged reference views. We propose a novel technique for detection
and identification of building facades from geo-tagged reference view using the
map and geometry of the building facades. We compute the likelihood of camera
location and orientation of the query images using the detected landmark
(building) identities from reference views, 2D map of the environment, and
geometry of building facades. We evaluate our approach for building
identification and geo-localization on a new challenging outdoors urban dataset
exhibiting large variations in appearance and viewpoint.
| new_dataset | 0.958187 |
1602.08751 | Jacopo Iacovacci | Jacopo Iacovacci and Ginestra Bianconi | Extracting Information from Multiplex Networks | 11 pages; 5 figures | Chaos: An Interdisciplinary Journal of Nonlinear Science 26.6
(2016): 065306 | 10.1063/1.4953161 | null | physics.soc-ph cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multiplex networks are generalized network structures that are able to
describe networks in which the same set of nodes are connected by links that
have different connotations. Multiplex networks are ubiquitous since they
describe social, financial, engineering and biological networks as well.
Extending our ability to analyze complex networks to multiplex network
structures increases greatly the level of information that is possible to
extract from Big Data. For these reasons characterizing the centrality of nodes
in multiplex networks and finding new ways to solve challenging inference
problems defined on multiplex networks are fundamental questions of network
science. In this paper we discuss the relevance of the Multiplex PageRank
algorithm for measuring the centrality of nodes in multilayer networks and we
characterize the utility of the recently introduced indicator function
$\widetilde{\Theta}^{S}$ for describing their mesoscale organization and
community structure. As working examples for studying these measures we
consider three multiplex network datasets coming for social science.
| [
{
"version": "v1",
"created": "Sun, 28 Feb 2016 18:32:14 GMT"
},
{
"version": "v2",
"created": "Mon, 16 May 2016 19:04:56 GMT"
}
] | 2016-09-01T00:00:00 | [
[
"Iacovacci",
"Jacopo",
""
],
[
"Bianconi",
"Ginestra",
""
]
] | TITLE: Extracting Information from Multiplex Networks
ABSTRACT: Multiplex networks are generalized network structures that are able to
describe networks in which the same set of nodes are connected by links that
have different connotations. Multiplex networks are ubiquitous since they
describe social, financial, engineering and biological networks as well.
Extending our ability to analyze complex networks to multiplex network
structures increases greatly the level of information that is possible to
extract from Big Data. For these reasons characterizing the centrality of nodes
in multiplex networks and finding new ways to solve challenging inference
problems defined on multiplex networks are fundamental questions of network
science. In this paper we discuss the relevance of the Multiplex PageRank
algorithm for measuring the centrality of nodes in multilayer networks and we
characterize the utility of the recently introduced indicator function
$\widetilde{\Theta}^{S}$ for describing their mesoscale organization and
community structure. As working examples for studying these measures we
consider three multiplex network datasets coming for social science.
| no_new_dataset | 0.953188 |
1606.01455 | Jin-Hwa Kim | Jin-Hwa Kim, Sang-Woo Lee, Dong-Hyun Kwak, Min-Oh Heo, Jeonghee Kim,
Jung-Woo Ha, Byoung-Tak Zhang | Multimodal Residual Learning for Visual QA | 13 pages, 7 figures, accepted for NIPS 2016 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep neural networks continue to advance the state-of-the-art of image
recognition tasks with various methods. However, applications of these methods
to multimodality remain limited. We present Multimodal Residual Networks (MRN)
for the multimodal residual learning of visual question-answering, which
extends the idea of the deep residual learning. Unlike the deep residual
learning, MRN effectively learns the joint representation from vision and
language information. The main idea is to use element-wise multiplication for
the joint residual mappings exploiting the residual learning of the attentional
models in recent studies. Various alternative models introduced by
multimodality are explored based on our study. We achieve the state-of-the-art
results on the Visual QA dataset for both Open-Ended and Multiple-Choice tasks.
Moreover, we introduce a novel method to visualize the attention effect of the
joint representations for each learning block using back-propagation algorithm,
even though the visual features are collapsed without spatial information.
| [
{
"version": "v1",
"created": "Sun, 5 Jun 2016 02:38:20 GMT"
},
{
"version": "v2",
"created": "Wed, 31 Aug 2016 08:28:38 GMT"
}
] | 2016-09-01T00:00:00 | [
[
"Kim",
"Jin-Hwa",
""
],
[
"Lee",
"Sang-Woo",
""
],
[
"Kwak",
"Dong-Hyun",
""
],
[
"Heo",
"Min-Oh",
""
],
[
"Kim",
"Jeonghee",
""
],
[
"Ha",
"Jung-Woo",
""
],
[
"Zhang",
"Byoung-Tak",
""
]
] | TITLE: Multimodal Residual Learning for Visual QA
ABSTRACT: Deep neural networks continue to advance the state-of-the-art of image
recognition tasks with various methods. However, applications of these methods
to multimodality remain limited. We present Multimodal Residual Networks (MRN)
for the multimodal residual learning of visual question-answering, which
extends the idea of the deep residual learning. Unlike the deep residual
learning, MRN effectively learns the joint representation from vision and
language information. The main idea is to use element-wise multiplication for
the joint residual mappings exploiting the residual learning of the attentional
models in recent studies. Various alternative models introduced by
multimodality are explored based on our study. We achieve the state-of-the-art
results on the Visual QA dataset for both Open-Ended and Multiple-Choice tasks.
Moreover, we introduce a novel method to visualize the attention effect of the
joint representations for each learning block using back-propagation algorithm,
even though the visual features are collapsed without spatial information.
| no_new_dataset | 0.9434 |
1608.08526 | Umar Iqbal | Umar Iqbal and Juergen Gall | Multi-Person Pose Estimation with Local Joint-to-Person Associations | Accepted to European Conference on Computer Vision (ECCV) Workshops,
Crowd Understanding, 2016 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite of the recent success of neural networks for human pose estimation,
current approaches are limited to pose estimation of a single person and cannot
handle humans in groups or crowds. In this work, we propose a method that
estimates the poses of multiple persons in an image in which a person can be
occluded by another person or might be truncated. To this end, we consider
multi-person pose estimation as a joint-to-person association problem. We
construct a fully connected graph from a set of detected joint candidates in an
image and resolve the joint-to-person association and outlier detection using
integer linear programming. Since solving joint-to-person association jointly
for all persons in an image is an NP-hard problem and even approximations are
expensive, we solve the problem locally for each person. On the challenging
MPII Human Pose Dataset for multiple persons, our approach achieves the
accuracy of a state-of-the-art method, but it is 6,000 to 19,000 times faster.
| [
{
"version": "v1",
"created": "Tue, 30 Aug 2016 16:00:42 GMT"
},
{
"version": "v2",
"created": "Wed, 31 Aug 2016 09:26:57 GMT"
}
] | 2016-09-01T00:00:00 | [
[
"Iqbal",
"Umar",
""
],
[
"Gall",
"Juergen",
""
]
] | TITLE: Multi-Person Pose Estimation with Local Joint-to-Person Associations
ABSTRACT: Despite of the recent success of neural networks for human pose estimation,
current approaches are limited to pose estimation of a single person and cannot
handle humans in groups or crowds. In this work, we propose a method that
estimates the poses of multiple persons in an image in which a person can be
occluded by another person or might be truncated. To this end, we consider
multi-person pose estimation as a joint-to-person association problem. We
construct a fully connected graph from a set of detected joint candidates in an
image and resolve the joint-to-person association and outlier detection using
integer linear programming. Since solving joint-to-person association jointly
for all persons in an image is an NP-hard problem and even approximations are
expensive, we solve the problem locally for each person. On the challenging
MPII Human Pose Dataset for multiple persons, our approach achieves the
accuracy of a state-of-the-art method, but it is 6,000 to 19,000 times faster.
| no_new_dataset | 0.9463 |
1608.08716 | Aishwarya Agrawal | C. Lawrence Zitnick, Aishwarya Agrawal, Stanislaw Antol, Margaret
Mitchell, Dhruv Batra, Devi Parikh | Measuring Machine Intelligence Through Visual Question Answering | AI Magazine, 2016 | null | null | null | cs.AI cs.CL cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As machines have become more intelligent, there has been a renewed interest
in methods for measuring their intelligence. A common approach is to propose
tasks for which a human excels, but one which machines find difficult. However,
an ideal task should also be easy to evaluate and not be easily gameable. We
begin with a case study exploring the recently popular task of image captioning
and its limitations as a task for measuring machine intelligence. An
alternative and more promising task is Visual Question Answering that tests a
machine's ability to reason about language and vision. We describe a dataset
unprecedented in size created for the task that contains over 760,000 human
generated questions about images. Using around 10 million human generated
answers, machines may be easily evaluated.
| [
{
"version": "v1",
"created": "Wed, 31 Aug 2016 02:56:00 GMT"
}
] | 2016-09-01T00:00:00 | [
[
"Zitnick",
"C. Lawrence",
""
],
[
"Agrawal",
"Aishwarya",
""
],
[
"Antol",
"Stanislaw",
""
],
[
"Mitchell",
"Margaret",
""
],
[
"Batra",
"Dhruv",
""
],
[
"Parikh",
"Devi",
""
]
] | TITLE: Measuring Machine Intelligence Through Visual Question Answering
ABSTRACT: As machines have become more intelligent, there has been a renewed interest
in methods for measuring their intelligence. A common approach is to propose
tasks for which a human excels, but one which machines find difficult. However,
an ideal task should also be easy to evaluate and not be easily gameable. We
begin with a case study exploring the recently popular task of image captioning
and its limitations as a task for measuring machine intelligence. An
alternative and more promising task is Visual Question Answering that tests a
machine's ability to reason about language and vision. We describe a dataset
unprecedented in size created for the task that contains over 760,000 human
generated questions about images. Using around 10 million human generated
answers, machines may be easily evaluated.
| new_dataset | 0.954393 |
1608.09002 | Prantik Bhattacharyya | Nemanja Spasojevic, Prantik Bhattacharyya, Adithya Rao | Mining Half a Billion Topical Experts Across Multiple Social Networks | 20 pages, 9 figures, 6 tables | null | 10.1007/s13278-016-0356-7 | null | cs.IR cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Mining topical experts on social media is a problem that has gained
significant attention due to its wide-ranging applications. Here we present the
first study that combines data from four major social networks -- Twitter,
Facebook, Google+ and LinkedIn, along with the Wikipedia graph and internet
webpage text and metadata, to rank topical experts across the global population
of users. We perform an in-depth analysis of 37 features derived from various
data sources such as message text, user lists, webpages, social graphs and
wikipedia. This large-scale study includes more than 12 billion messages over a
90-day sliding window and 58 billion social graph edges. Comparison reveals
that features derived from Twitter Lists, Wikipedia, internet webpages and
Twitter Followers are especially good indicators of expertise. We train an
expertise ranking model using these features on a large ground truth dataset
containing almost 90,000 labels. This model is applied within a production
system that ranks over 650 million experts in more than 9,000 topical domains
on a daily basis. We provide results and examples on the effectiveness of our
expert ranking system, along with empirical validation. Finally, we make the
topical expertise data available through open REST APIs for wider use.
| [
{
"version": "v1",
"created": "Wed, 31 Aug 2016 19:14:03 GMT"
}
] | 2016-09-01T00:00:00 | [
[
"Spasojevic",
"Nemanja",
""
],
[
"Bhattacharyya",
"Prantik",
""
],
[
"Rao",
"Adithya",
""
]
] | TITLE: Mining Half a Billion Topical Experts Across Multiple Social Networks
ABSTRACT: Mining topical experts on social media is a problem that has gained
significant attention due to its wide-ranging applications. Here we present the
first study that combines data from four major social networks -- Twitter,
Facebook, Google+ and LinkedIn, along with the Wikipedia graph and internet
webpage text and metadata, to rank topical experts across the global population
of users. We perform an in-depth analysis of 37 features derived from various
data sources such as message text, user lists, webpages, social graphs and
wikipedia. This large-scale study includes more than 12 billion messages over a
90-day sliding window and 58 billion social graph edges. Comparison reveals
that features derived from Twitter Lists, Wikipedia, internet webpages and
Twitter Followers are especially good indicators of expertise. We train an
expertise ranking model using these features on a large ground truth dataset
containing almost 90,000 labels. This model is applied within a production
system that ranks over 650 million experts in more than 9,000 topical domains
on a daily basis. We provide results and examples on the effectiveness of our
expert ranking system, along with empirical validation. Finally, we make the
topical expertise data available through open REST APIs for wider use.
| new_dataset | 0.828037 |
1604.06506 | Roeland De Geest | Roeland De Geest, Efstratios Gavves, Amir Ghodrati, Zhenyang Li, Cees
Snoek, Tinne Tuytelaars | Online Action Detection | Project page:
http://homes.esat.kuleuven.be/~rdegeest/OnlineActionDetection.html | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In online action detection, the goal is to detect the start of an action in a
video stream as soon as it happens. For instance, if a child is chasing a ball,
an autonomous car should recognize what is going on and respond immediately.
This is a very challenging problem for four reasons. First, only partial
actions are observed. Second, there is a large variability in negative data.
Third, the start of the action is unknown, so it is unclear over what time
window the information should be integrated. Finally, in real world data, large
within-class variability exists. This problem has been addressed before, but
only to some extent. Our contributions to online action detection are
threefold. First, we introduce a realistic dataset composed of 27 episodes from
6 popular TV series. The dataset spans over 16 hours of footage annotated with
30 action classes, totaling 6,231 action instances. Second, we analyze and
compare various baseline methods, showing this is a challenging problem for
which none of the methods provides a good solution. Third, we analyze the
change in performance when there is a variation in viewpoint, occlusion,
truncation, etc. We introduce an evaluation protocol for fair comparison. The
dataset, the baselines and the models will all be made publicly available to
encourage (much needed) further research on online action detection on
realistic data.
| [
{
"version": "v1",
"created": "Thu, 21 Apr 2016 22:02:50 GMT"
},
{
"version": "v2",
"created": "Tue, 30 Aug 2016 09:29:39 GMT"
}
] | 2016-08-31T00:00:00 | [
[
"De Geest",
"Roeland",
""
],
[
"Gavves",
"Efstratios",
""
],
[
"Ghodrati",
"Amir",
""
],
[
"Li",
"Zhenyang",
""
],
[
"Snoek",
"Cees",
""
],
[
"Tuytelaars",
"Tinne",
""
]
] | TITLE: Online Action Detection
ABSTRACT: In online action detection, the goal is to detect the start of an action in a
video stream as soon as it happens. For instance, if a child is chasing a ball,
an autonomous car should recognize what is going on and respond immediately.
This is a very challenging problem for four reasons. First, only partial
actions are observed. Second, there is a large variability in negative data.
Third, the start of the action is unknown, so it is unclear over what time
window the information should be integrated. Finally, in real world data, large
within-class variability exists. This problem has been addressed before, but
only to some extent. Our contributions to online action detection are
threefold. First, we introduce a realistic dataset composed of 27 episodes from
6 popular TV series. The dataset spans over 16 hours of footage annotated with
30 action classes, totaling 6,231 action instances. Second, we analyze and
compare various baseline methods, showing this is a challenging problem for
which none of the methods provides a good solution. Third, we analyze the
change in performance when there is a variation in viewpoint, occlusion,
truncation, etc. We introduce an evaluation protocol for fair comparison. The
dataset, the baselines and the models will all be made publicly available to
encourage (much needed) further research on online action detection on
realistic data.
| new_dataset | 0.965803 |
1608.07550 | Joyce Fang | Joyce Fang, Dmitry Savransky | Automated alignment of a reconfigurable optical system using focal-plane
sensing and Kalman filtering | null | null | 10.1364/AO.55.005967 | null | astro-ph.IM physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Automation of alignment tasks can provide improved efficiency and greatly
increase the flexibility of an optical system. Current optical systems with
automated alignment capabilities are typically designed to include a dedicated
wavefront sensor. Here, we demonstrate a self-aligning method for a
reconfigurable system using only focal plane images. We define a two lens
optical system with eight degrees of freedom. Images are simulated given
misalignment parameters using ZEMAX software. We perform a principal component
analysis (PCA) on the simulated dataset to obtain Karhunen-Lo\`eve (KL) modes,
which form the basis set whose weights are the system measurements. A model
function which maps the state to the measurement is learned using nonlinear
least squares fitting and serves as the measurement function for the nonlinear
estimator (Extended and Unscented Kalman filters) used to calculate control
inputs to align the system. We present and discuss both simulated and
experimental results of the full system in operation.
| [
{
"version": "v1",
"created": "Fri, 26 Aug 2016 18:24:18 GMT"
}
] | 2016-08-31T00:00:00 | [
[
"Fang",
"Joyce",
""
],
[
"Savransky",
"Dmitry",
""
]
] | TITLE: Automated alignment of a reconfigurable optical system using focal-plane
sensing and Kalman filtering
ABSTRACT: Automation of alignment tasks can provide improved efficiency and greatly
increase the flexibility of an optical system. Current optical systems with
automated alignment capabilities are typically designed to include a dedicated
wavefront sensor. Here, we demonstrate a self-aligning method for a
reconfigurable system using only focal plane images. We define a two lens
optical system with eight degrees of freedom. Images are simulated given
misalignment parameters using ZEMAX software. We perform a principal component
analysis (PCA) on the simulated dataset to obtain Karhunen-Lo\`eve (KL) modes,
which form the basis set whose weights are the system measurements. A model
function which maps the state to the measurement is learned using nonlinear
least squares fitting and serves as the measurement function for the nonlinear
estimator (Extended and Unscented Kalman filters) used to calculate control
inputs to align the system. We present and discuss both simulated and
experimental results of the full system in operation.
| no_new_dataset | 0.947769 |
1608.08148 | Olaf Hartig | Olaf Hartig and Carlos Buil-Aranda | brTPF: Bindings-Restricted Triple Pattern Fragments (Extended Preprint) | This document is an extended preprint of a paper published in the
proceedings of the ODBASE 2016 conference. In contrast to the proceedings
version, this document contains Appendixes A and B which present additional
experimental results | null | null | null | cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Triple Pattern Fragment (TPF) interface is a recent proposal for reducing
server load in Web-based approaches to execute SPARQL queries over public RDF
datasets. The price for less overloaded servers is a higher client-side load
and a substantial increase in network load (in terms of both the number of HTTP
requests and data transfer). In this paper, we propose a slightly extended
interface that allows clients to attach intermediate results to triple pattern
requests. The response to such a request is expected to contain triples from
the underlying dataset that do not only match the given triple pattern (as in
the case of TPF), but that are guaranteed to contribute in a join with the
given intermediate result. Our hypothesis is that a distributed query execution
using this extended interface can reduce the network load (in comparison to a
pure TPF-based query execution) without reducing the overall throughput of the
client-server system significantly. Our main contribution in this paper is
twofold: we empirically verify the hypothesis and provide an extensive
experimental comparison of our proposal and TPF.
| [
{
"version": "v1",
"created": "Mon, 29 Aug 2016 17:09:53 GMT"
},
{
"version": "v2",
"created": "Tue, 30 Aug 2016 15:49:29 GMT"
}
] | 2016-08-31T00:00:00 | [
[
"Hartig",
"Olaf",
""
],
[
"Buil-Aranda",
"Carlos",
""
]
] | TITLE: brTPF: Bindings-Restricted Triple Pattern Fragments (Extended Preprint)
ABSTRACT: The Triple Pattern Fragment (TPF) interface is a recent proposal for reducing
server load in Web-based approaches to execute SPARQL queries over public RDF
datasets. The price for less overloaded servers is a higher client-side load
and a substantial increase in network load (in terms of both the number of HTTP
requests and data transfer). In this paper, we propose a slightly extended
interface that allows clients to attach intermediate results to triple pattern
requests. The response to such a request is expected to contain triples from
the underlying dataset that do not only match the given triple pattern (as in
the case of TPF), but that are guaranteed to contribute in a join with the
given intermediate result. Our hypothesis is that a distributed query execution
using this extended interface can reduce the network load (in comparison to a
pure TPF-based query execution) without reducing the overall throughput of the
client-server system significantly. Our main contribution in this paper is
twofold: we empirically verify the hypothesis and provide an extensive
experimental comparison of our proposal and TPF.
| no_new_dataset | 0.948917 |
1608.08242 | Colin Lea | Colin Lea, Rene Vidal, Austin Reiter, Gregory D. Hager | Temporal Convolutional Networks: A Unified Approach to Action
Segmentation | Submitted to the ECCV workshop on "Brave new ideas for motion
representations in videos" (http://bravenewmotion.github.io/) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The dominant paradigm for video-based action segmentation is composed of two
steps: first, for each frame, compute low-level features using Dense
Trajectories or a Convolutional Neural Network that encode spatiotemporal
information locally, and second, input these features into a classifier that
captures high-level temporal relationships, such as a Recurrent Neural Network
(RNN). While often effective, this decoupling requires specifying two separate
models, each with their own complexities, and prevents capturing more nuanced
long-range spatiotemporal relationships. We propose a unified approach, as
demonstrated by our Temporal Convolutional Network (TCN), that hierarchically
captures relationships at low-, intermediate-, and high-level time-scales. Our
model achieves superior or competitive performance using video or sensor data
on three public action segmentation datasets and can be trained in a fraction
of the time it takes to train an RNN.
| [
{
"version": "v1",
"created": "Mon, 29 Aug 2016 20:48:15 GMT"
}
] | 2016-08-31T00:00:00 | [
[
"Lea",
"Colin",
""
],
[
"Vidal",
"Rene",
""
],
[
"Reiter",
"Austin",
""
],
[
"Hager",
"Gregory D.",
""
]
] | TITLE: Temporal Convolutional Networks: A Unified Approach to Action
Segmentation
ABSTRACT: The dominant paradigm for video-based action segmentation is composed of two
steps: first, for each frame, compute low-level features using Dense
Trajectories or a Convolutional Neural Network that encode spatiotemporal
information locally, and second, input these features into a classifier that
captures high-level temporal relationships, such as a Recurrent Neural Network
(RNN). While often effective, this decoupling requires specifying two separate
models, each with their own complexities, and prevents capturing more nuanced
long-range spatiotemporal relationships. We propose a unified approach, as
demonstrated by our Temporal Convolutional Network (TCN), that hierarchically
captures relationships at low-, intermediate-, and high-level time-scales. Our
model achieves superior or competitive performance using video or sensor data
on three public action segmentation datasets and can be trained in a fraction
of the time it takes to train an RNN.
| no_new_dataset | 0.950134 |
1608.08305 | Ronghang Hu | Ronghang Hu, Marcus Rohrbach, Subhashini Venugopalan, Trevor Darrell | Utilizing Large Scale Vision and Text Datasets for Image Segmentation
from Referring Expressions | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Image segmentation from referring expressions is a joint vision and language
modeling task, where the input is an image and a textual expression describing
a particular region in the image; and the goal is to localize and segment the
specific image region based on the given expression. One major difficulty to
train such language-based image segmentation systems is the lack of datasets
with joint vision and text annotations. Although existing vision datasets such
as MS COCO provide image captions, there are few datasets with region-level
textual annotations for images, and these are often smaller in scale. In this
paper, we explore how existing large scale vision-only and text-only datasets
can be utilized to train models for image segmentation from referring
expressions. We propose a method to address this problem, and show in
experiments that our method can help this joint vision and language modeling
task with vision-only and text-only data and outperforms previous results.
| [
{
"version": "v1",
"created": "Tue, 30 Aug 2016 02:27:41 GMT"
}
] | 2016-08-31T00:00:00 | [
[
"Hu",
"Ronghang",
""
],
[
"Rohrbach",
"Marcus",
""
],
[
"Venugopalan",
"Subhashini",
""
],
[
"Darrell",
"Trevor",
""
]
] | TITLE: Utilizing Large Scale Vision and Text Datasets for Image Segmentation
from Referring Expressions
ABSTRACT: Image segmentation from referring expressions is a joint vision and language
modeling task, where the input is an image and a textual expression describing
a particular region in the image; and the goal is to localize and segment the
specific image region based on the given expression. One major difficulty to
train such language-based image segmentation systems is the lack of datasets
with joint vision and text annotations. Although existing vision datasets such
as MS COCO provide image captions, there are few datasets with region-level
textual annotations for images, and these are often smaller in scale. In this
paper, we explore how existing large scale vision-only and text-only datasets
can be utilized to train models for image segmentation from referring
expressions. We propose a method to address this problem, and show in
experiments that our method can help this joint vision and language modeling
task with vision-only and text-only data and outperforms previous results.
| no_new_dataset | 0.949856 |
1608.08471 | Johannes Stegmaier | Johannes Stegmaier | New Methods to Improve Large-Scale Microscopy Image Analysis with Prior
Knowledge and Uncertainty | 218 pages, 58 figures, PhD thesis, Department of Mechanical
Engineering, Karlsruhe Institute of Technology, published online with KITopen
(License: CC BY-SA 3.0, http://dx.doi.org/10.5445/IR/1000057821) | null | 10.5445/IR/1000057821 | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Multidimensional imaging techniques provide powerful ways to examine various
kinds of scientific questions. The routinely produced datasets in the
terabyte-range, however, can hardly be analyzed manually and require an
extensive use of automated image analysis. The present thesis introduces a new
concept for the estimation and propagation of uncertainty involved in image
analysis operators and new segmentation algorithms that are suitable for
terabyte-scale analyses of 3D+t microscopy images.
| [
{
"version": "v1",
"created": "Tue, 30 Aug 2016 14:21:55 GMT"
}
] | 2016-08-31T00:00:00 | [
[
"Stegmaier",
"Johannes",
""
]
] | TITLE: New Methods to Improve Large-Scale Microscopy Image Analysis with Prior
Knowledge and Uncertainty
ABSTRACT: Multidimensional imaging techniques provide powerful ways to examine various
kinds of scientific questions. The routinely produced datasets in the
terabyte-range, however, can hardly be analyzed manually and require an
extensive use of automated image analysis. The present thesis introduces a new
concept for the estimation and propagation of uncertainty involved in image
analysis operators and new segmentation algorithms that are suitable for
terabyte-scale analyses of 3D+t microscopy images.
| no_new_dataset | 0.944791 |
1511.04599 | Seyed-Mohsen Moosavi-Dezfooli | Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Pascal Frossard | DeepFool: a simple and accurate method to fool deep neural networks | In Proceedings of IEEE Conference on Computer Vision and Pattern
Recognition (CVPR), 2016 | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | State-of-the-art deep neural networks have achieved impressive results on
many image classification tasks. However, these same architectures have been
shown to be unstable to small, well sought, perturbations of the images.
Despite the importance of this phenomenon, no effective methods have been
proposed to accurately compute the robustness of state-of-the-art deep
classifiers to such perturbations on large-scale datasets. In this paper, we
fill this gap and propose the DeepFool algorithm to efficiently compute
perturbations that fool deep networks, and thus reliably quantify the
robustness of these classifiers. Extensive experimental results show that our
approach outperforms recent methods in the task of computing adversarial
perturbations and making classifiers more robust.
| [
{
"version": "v1",
"created": "Sat, 14 Nov 2015 18:50:00 GMT"
},
{
"version": "v2",
"created": "Thu, 14 Apr 2016 09:33:23 GMT"
},
{
"version": "v3",
"created": "Mon, 4 Jul 2016 04:49:44 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Moosavi-Dezfooli",
"Seyed-Mohsen",
""
],
[
"Fawzi",
"Alhussein",
""
],
[
"Frossard",
"Pascal",
""
]
] | TITLE: DeepFool: a simple and accurate method to fool deep neural networks
ABSTRACT: State-of-the-art deep neural networks have achieved impressive results on
many image classification tasks. However, these same architectures have been
shown to be unstable to small, well sought, perturbations of the images.
Despite the importance of this phenomenon, no effective methods have been
proposed to accurately compute the robustness of state-of-the-art deep
classifiers to such perturbations on large-scale datasets. In this paper, we
fill this gap and propose the DeepFool algorithm to efficiently compute
perturbations that fool deep networks, and thus reliably quantify the
robustness of these classifiers. Extensive experimental results show that our
approach outperforms recent methods in the task of computing adversarial
perturbations and making classifiers more robust.
| no_new_dataset | 0.947088 |
1601.06579 | Dong Nguyen | Dong Nguyen, Jacob Eisenstein | A Kernel Independence Test for Geographical Language Variation | In submission. 26 pages | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantifying the degree of spatial dependence for linguistic variables is a
key task for analyzing dialectal variation. However, existing approaches have
important drawbacks. First, they are based on parametric models of dependence,
which limits their power in cases where the underlying parametric assumptions
are violated. Second, they are not applicable to all types of linguistic data:
some approaches apply only to frequencies, others to boolean indicators of
whether a linguistic variable is present. We present a new method for measuring
geographical language variation, which solves both of these problems. Our
approach builds on Reproducing Kernel Hilbert space (RKHS) representations for
nonparametric statistics, and takes the form of a test statistic that is
computed from pairs of individual geotagged observations without aggregation
into predefined geographical bins. We compare this test with prior work using
synthetic data as well as a diverse set of real datasets: a corpus of Dutch
tweets, a Dutch syntactic atlas, and a dataset of letters to the editor in
North American newspapers. Our proposed test is shown to support robust
inferences across a broad range of scenarios and types of data.
| [
{
"version": "v1",
"created": "Mon, 25 Jan 2016 12:45:59 GMT"
},
{
"version": "v2",
"created": "Mon, 29 Aug 2016 13:16:42 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Nguyen",
"Dong",
""
],
[
"Eisenstein",
"Jacob",
""
]
] | TITLE: A Kernel Independence Test for Geographical Language Variation
ABSTRACT: Quantifying the degree of spatial dependence for linguistic variables is a
key task for analyzing dialectal variation. However, existing approaches have
important drawbacks. First, they are based on parametric models of dependence,
which limits their power in cases where the underlying parametric assumptions
are violated. Second, they are not applicable to all types of linguistic data:
some approaches apply only to frequencies, others to boolean indicators of
whether a linguistic variable is present. We present a new method for measuring
geographical language variation, which solves both of these problems. Our
approach builds on Reproducing Kernel Hilbert space (RKHS) representations for
nonparametric statistics, and takes the form of a test statistic that is
computed from pairs of individual geotagged observations without aggregation
into predefined geographical bins. We compare this test with prior work using
synthetic data as well as a diverse set of real datasets: a corpus of Dutch
tweets, a Dutch syntactic atlas, and a dataset of letters to the editor in
North American newspapers. Our proposed test is shown to support robust
inferences across a broad range of scenarios and types of data.
| no_new_dataset | 0.92944 |
1605.04457 | Alexandre Mauroy | Alexandre Mauroy and Jorge Goncalves | Linear identification of nonlinear systems: A lifting technique based on
the Koopman operator | 6 pages | null | null | null | cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We exploit the key idea that nonlinear system identification is equivalent to
linear identification of the socalled Koopman operator. Instead of considering
nonlinear system identification in the state space, we obtain a novel linear
identification technique by recasting the problem in the infinite-dimensional
space of observables. This technique can be described in two main steps. In the
first step, similar to the socalled Extended Dynamic Mode Decomposition
algorithm, the data are lifted to the infinite-dimensional space and used for
linear identification of the Koopman operator. In the second step, the obtained
Koopman operator is "projected back" to the finite-dimensional state space, and
identified to the nonlinear vector field through a linear least squares
problem. The proposed technique is efficient to recover (polynomial) vector
fields of different classes of systems, including unstable, chaotic, and open
systems. In addition, it is robust to noise, well-suited to model low sampling
rate datasets, and able to infer network topology and dynamics.
| [
{
"version": "v1",
"created": "Sat, 14 May 2016 19:05:18 GMT"
},
{
"version": "v2",
"created": "Sat, 27 Aug 2016 18:21:39 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Mauroy",
"Alexandre",
""
],
[
"Goncalves",
"Jorge",
""
]
] | TITLE: Linear identification of nonlinear systems: A lifting technique based on
the Koopman operator
ABSTRACT: We exploit the key idea that nonlinear system identification is equivalent to
linear identification of the socalled Koopman operator. Instead of considering
nonlinear system identification in the state space, we obtain a novel linear
identification technique by recasting the problem in the infinite-dimensional
space of observables. This technique can be described in two main steps. In the
first step, similar to the socalled Extended Dynamic Mode Decomposition
algorithm, the data are lifted to the infinite-dimensional space and used for
linear identification of the Koopman operator. In the second step, the obtained
Koopman operator is "projected back" to the finite-dimensional state space, and
identified to the nonlinear vector field through a linear least squares
problem. The proposed technique is efficient to recover (polynomial) vector
fields of different classes of systems, including unstable, chaotic, and open
systems. In addition, it is robust to noise, well-suited to model low sampling
rate datasets, and able to infer network topology and dynamics.
| no_new_dataset | 0.948106 |
1606.00850 | Tianfu Wu | Yunzhu Li, Benyuan Sun, Tianfu Wu and Yizhou Wang | Face Detection with End-to-End Integration of a ConvNet and a 3D Model | 16 pages, Y. Li and B. Sun contributed equally to this work | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents a method for face detection in the wild, which integrates
a ConvNet and a 3D mean face model in an end-to-end multi-task discriminative
learning framework. The 3D mean face model is predefined and fixed (e.g., we
used the one provided in the AFLW dataset). The ConvNet consists of two
components: (i) The face pro- posal component computes face bounding box
proposals via estimating facial key-points and the 3D transformation (rotation
and translation) parameters for each predicted key-point w.r.t. the 3D mean
face model. (ii) The face verification component computes detection results by
prun- ing and refining proposals based on facial key-points based configuration
pooling. The proposed method addresses two issues in adapting state- of-the-art
generic object detection ConvNets (e.g., faster R-CNN) for face detection: (i)
One is to eliminate the heuristic design of prede- fined anchor boxes in the
region proposals network (RPN) by exploit- ing a 3D mean face model. (ii) The
other is to replace the generic RoI (Region-of-Interest) pooling layer with a
configuration pooling layer to respect underlying object structures. The
multi-task loss consists of three terms: the classification Softmax loss and
the location smooth l1 -losses [14] of both the facial key-points and the face
bounding boxes. In ex- periments, our ConvNet is trained on the AFLW dataset
only and tested on the FDDB benchmark with fine-tuning and on the AFW benchmark
without fine-tuning. The proposed method obtains very competitive
state-of-the-art performance in the two benchmarks.
| [
{
"version": "v1",
"created": "Thu, 2 Jun 2016 20:08:28 GMT"
},
{
"version": "v2",
"created": "Tue, 26 Jul 2016 03:49:44 GMT"
},
{
"version": "v3",
"created": "Mon, 29 Aug 2016 14:57:17 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Li",
"Yunzhu",
""
],
[
"Sun",
"Benyuan",
""
],
[
"Wu",
"Tianfu",
""
],
[
"Wang",
"Yizhou",
""
]
] | TITLE: Face Detection with End-to-End Integration of a ConvNet and a 3D Model
ABSTRACT: This paper presents a method for face detection in the wild, which integrates
a ConvNet and a 3D mean face model in an end-to-end multi-task discriminative
learning framework. The 3D mean face model is predefined and fixed (e.g., we
used the one provided in the AFLW dataset). The ConvNet consists of two
components: (i) The face pro- posal component computes face bounding box
proposals via estimating facial key-points and the 3D transformation (rotation
and translation) parameters for each predicted key-point w.r.t. the 3D mean
face model. (ii) The face verification component computes detection results by
prun- ing and refining proposals based on facial key-points based configuration
pooling. The proposed method addresses two issues in adapting state- of-the-art
generic object detection ConvNets (e.g., faster R-CNN) for face detection: (i)
One is to eliminate the heuristic design of prede- fined anchor boxes in the
region proposals network (RPN) by exploit- ing a 3D mean face model. (ii) The
other is to replace the generic RoI (Region-of-Interest) pooling layer with a
configuration pooling layer to respect underlying object structures. The
multi-task loss consists of three terms: the classification Softmax loss and
the location smooth l1 -losses [14] of both the facial key-points and the face
bounding boxes. In ex- periments, our ConvNet is trained on the AFLW dataset
only and tested on the FDDB benchmark with fine-tuning and on the AFW benchmark
without fine-tuning. The proposed method obtains very competitive
state-of-the-art performance in the two benchmarks.
| no_new_dataset | 0.952309 |
1607.00669 | Charbel Sakr | Charbel Sakr, Ameya Patil, Sai Zhang, Yongjune Kim, Naresh Shanbhag | Understanding the Energy and Precision Requirements for Online Learning | 14 pages, 5 figures 4 of which have 2 subfigures | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It is well-known that the precision of data, hyperparameters, and internal
representations employed in learning systems directly impacts its energy,
throughput, and latency. The precision requirements for the training algorithm
are also important for systems that learn on-the-fly. Prior work has shown that
the data and hyperparameters can be quantized heavily without incurring much
penalty in classification accuracy when compared to floating point
implementations. These works suffer from two key limitations. First, they
assume uniform precision for the classifier and for the training algorithm and
thus miss out on the opportunity to further reduce precision. Second, prior
works are empirical studies. In this article, we overcome both these
limitations by deriving analytical lower bounds on the precision requirements
of the commonly employed stochastic gradient descent (SGD) on-line learning
algorithm in the specific context of a support vector machine (SVM). Lower
bounds on the data precision are derived in terms of the the desired
classification accuracy and precision of the hyperparameters used in the
classifier. Additionally, lower bounds on the hyperparameter precision in the
SGD training algorithm are obtained. These bounds are validated using both
synthetic and the UCI breast cancer dataset. Additionally, the impact of these
precisions on the energy consumption of a fixed-point SVM with on-line training
is studied.
| [
{
"version": "v1",
"created": "Sun, 3 Jul 2016 18:54:25 GMT"
},
{
"version": "v2",
"created": "Mon, 15 Aug 2016 23:01:12 GMT"
},
{
"version": "v3",
"created": "Fri, 26 Aug 2016 21:56:03 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Sakr",
"Charbel",
""
],
[
"Patil",
"Ameya",
""
],
[
"Zhang",
"Sai",
""
],
[
"Kim",
"Yongjune",
""
],
[
"Shanbhag",
"Naresh",
""
]
] | TITLE: Understanding the Energy and Precision Requirements for Online Learning
ABSTRACT: It is well-known that the precision of data, hyperparameters, and internal
representations employed in learning systems directly impacts its energy,
throughput, and latency. The precision requirements for the training algorithm
are also important for systems that learn on-the-fly. Prior work has shown that
the data and hyperparameters can be quantized heavily without incurring much
penalty in classification accuracy when compared to floating point
implementations. These works suffer from two key limitations. First, they
assume uniform precision for the classifier and for the training algorithm and
thus miss out on the opportunity to further reduce precision. Second, prior
works are empirical studies. In this article, we overcome both these
limitations by deriving analytical lower bounds on the precision requirements
of the commonly employed stochastic gradient descent (SGD) on-line learning
algorithm in the specific context of a support vector machine (SVM). Lower
bounds on the data precision are derived in terms of the the desired
classification accuracy and precision of the hyperparameters used in the
classifier. Additionally, lower bounds on the hyperparameter precision in the
SGD training algorithm are obtained. These bounds are validated using both
synthetic and the UCI breast cancer dataset. Additionally, the impact of these
precisions on the energy consumption of a fixed-point SVM with on-line training
is studied.
| no_new_dataset | 0.952706 |
1608.06148 | Chandrajit M | Chandrajit M, Girisha R and Vasudev T | Multiple objects tracking in surveillance video using color and Hu
moments | 13 pages, Signal & Image Processing : An International Journal
(SIPIJ) Vol.7, No.3, June 2016 | null | 10.5121/sipij.2016.7302 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multiple objects tracking finds its applications in many high level vision
analysis like object behaviour interpretation and gait recognition. In this
paper, a feature based method to track the multiple moving objects in
surveillance video sequence is proposed. Object tracking is done by extracting
the color and Hu moments features from the motion segmented object blob and
establishing the association of objects in the successive frames of the video
sequence based on Chi-Square dissimilarity measure and nearest neighbor
classifier. The benchmark IEEE PETS and IEEE Change Detection datasets has been
used to show the robustness of the proposed method. The proposed method is
assessed quantitatively using the precision and recall accuracy metrics.
Further, comparative evaluation with related works has been carried out to
exhibit the efficacy of the proposed method.
| [
{
"version": "v1",
"created": "Mon, 22 Aug 2016 12:42:46 GMT"
},
{
"version": "v2",
"created": "Sun, 28 Aug 2016 13:11:35 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"M",
"Chandrajit",
""
],
[
"R",
"Girisha",
""
],
[
"T",
"Vasudev",
""
]
] | TITLE: Multiple objects tracking in surveillance video using color and Hu
moments
ABSTRACT: Multiple objects tracking finds its applications in many high level vision
analysis like object behaviour interpretation and gait recognition. In this
paper, a feature based method to track the multiple moving objects in
surveillance video sequence is proposed. Object tracking is done by extracting
the color and Hu moments features from the motion segmented object blob and
establishing the association of objects in the successive frames of the video
sequence based on Chi-Square dissimilarity measure and nearest neighbor
classifier. The benchmark IEEE PETS and IEEE Change Detection datasets has been
used to show the robustness of the proposed method. The proposed method is
assessed quantitatively using the precision and recall accuracy metrics.
Further, comparative evaluation with related works has been carried out to
exhibit the efficacy of the proposed method.
| no_new_dataset | 0.954858 |
1608.06347 | Shikai Jin | Shikai Jin, Yuxuan Cui, Chunli Yu | A New Parallelization Method for K-means | null | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | K-means is a popular clustering method used in data mining area. To work with
large datasets, researchers propose PKMeans, which is a parallel k-means on
MapReduce. However, the existing k-means parallelization methods including
PKMeans have many limitations. PKMeans can't finish all its iterations in one
MapReduce job, so it has to repeat cascading MapReduce jobs in a loop until
convergence. On the most popular MapReduce platform, Hadoop, every MapReduce
job introduces significant I/O overheads and extra execution time at stages of
job start-up and shuffling. Even worse, it has been proved that in the worst
case, k-means needs $2^{{\Omega}(n)}$ MapReduce jobs to converge, where n is
the number of data instances, which means huge overheads for large datasets.
Additionally, in PKMeans, at most one reducer can be assigned to and update
each centroid, so PKMeans can only make use of limited number of parallel
reducers. In this paper, we propose an improved parallel method for k-means,
IPKMeans, which has a parallel preprocessing stage using k-d tree and can
finish k-means in one single MapReduce job with much more reducers working in
parallel and lower I/O overheads than PKMeans and has a fast post-processing
stage generating the final result. In our method, both k-d tree and the new
improved parallel k-means are implemented using MapReduce and tested on Hadoop.
Our experiments show that with same dataset and initial centroids, our method
has up to 2/3 lower I/O overheads and consumes less amount of time than PKMeans
to get a very close clustering result.
| [
{
"version": "v1",
"created": "Tue, 23 Aug 2016 00:35:10 GMT"
},
{
"version": "v2",
"created": "Fri, 26 Aug 2016 21:11:34 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Jin",
"Shikai",
""
],
[
"Cui",
"Yuxuan",
""
],
[
"Yu",
"Chunli",
""
]
] | TITLE: A New Parallelization Method for K-means
ABSTRACT: K-means is a popular clustering method used in data mining area. To work with
large datasets, researchers propose PKMeans, which is a parallel k-means on
MapReduce. However, the existing k-means parallelization methods including
PKMeans have many limitations. PKMeans can't finish all its iterations in one
MapReduce job, so it has to repeat cascading MapReduce jobs in a loop until
convergence. On the most popular MapReduce platform, Hadoop, every MapReduce
job introduces significant I/O overheads and extra execution time at stages of
job start-up and shuffling. Even worse, it has been proved that in the worst
case, k-means needs $2^{{\Omega}(n)}$ MapReduce jobs to converge, where n is
the number of data instances, which means huge overheads for large datasets.
Additionally, in PKMeans, at most one reducer can be assigned to and update
each centroid, so PKMeans can only make use of limited number of parallel
reducers. In this paper, we propose an improved parallel method for k-means,
IPKMeans, which has a parallel preprocessing stage using k-d tree and can
finish k-means in one single MapReduce job with much more reducers working in
parallel and lower I/O overheads than PKMeans and has a fast post-processing
stage generating the final result. In our method, both k-d tree and the new
improved parallel k-means are implemented using MapReduce and tested on Hadoop.
Our experiments show that with same dataset and initial centroids, our method
has up to 2/3 lower I/O overheads and consumes less amount of time than PKMeans
to get a very close clustering result.
| no_new_dataset | 0.94625 |
1608.07636 | Hossein Hosseini | Hossein Hosseini, Sreeram Kannan, Baosen Zhang and Radha Poovendran | Learning Temporal Dependence from Time-Series Data with Latent Variables | null | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the setting where a collection of time series, modeled as random
processes, evolve in a causal manner, and one is interested in learning the
graph governing the relationships of these processes. A special case of wide
interest and applicability is the setting where the noise is Gaussian and
relationships are Markov and linear. We study this setting with two additional
features: firstly, each random process has a hidden (latent) state, which we
use to model the internal memory possessed by the variables (similar to hidden
Markov models). Secondly, each variable can depend on its latent memory state
through a random lag (rather than a fixed lag), thus modeling memory recall
with differing lags at distinct times. Under this setting, we develop an
estimator and prove that under a genericity assumption, the parameters of the
model can be learned consistently. We also propose a practical adaption of this
estimator, which demonstrates significant performance gains in both synthetic
and real-world datasets.
| [
{
"version": "v1",
"created": "Sat, 27 Aug 2016 00:25:54 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Hosseini",
"Hossein",
""
],
[
"Kannan",
"Sreeram",
""
],
[
"Zhang",
"Baosen",
""
],
[
"Poovendran",
"Radha",
""
]
] | TITLE: Learning Temporal Dependence from Time-Series Data with Latent Variables
ABSTRACT: We consider the setting where a collection of time series, modeled as random
processes, evolve in a causal manner, and one is interested in learning the
graph governing the relationships of these processes. A special case of wide
interest and applicability is the setting where the noise is Gaussian and
relationships are Markov and linear. We study this setting with two additional
features: firstly, each random process has a hidden (latent) state, which we
use to model the internal memory possessed by the variables (similar to hidden
Markov models). Secondly, each variable can depend on its latent memory state
through a random lag (rather than a fixed lag), thus modeling memory recall
with differing lags at distinct times. Under this setting, we develop an
estimator and prove that under a genericity assumption, the parameters of the
model can be learned consistently. We also propose a practical adaption of this
estimator, which demonstrates significant performance gains in both synthetic
and real-world datasets.
| no_new_dataset | 0.949059 |
1608.07639 | Yuval Atzmon | Yuval Atzmon, Jonathan Berant, Vahid Kezami, Amir Globerson and Gal
Chechik | Learning to generalize to new compositions in image understanding | null | null | null | null | cs.CV cs.AI cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recurrent neural networks have recently been used for learning to describe
images using natural language. However, it has been observed that these models
generalize poorly to scenes that were not observed during training, possibly
depending too strongly on the statistics of the text in the training data. Here
we propose to describe images using short structured representations, aiming to
capture the crux of a description. These structured representations allow us to
tease-out and evaluate separately two types of generalization: standard
generalization to new images with similar scenes, and generalization to new
combinations of known entities. We compare two learning approaches on the
MS-COCO dataset: a state-of-the-art recurrent network based on an LSTM (Show,
Attend and Tell), and a simple structured prediction model on top of a deep
network. We find that the structured model generalizes to new compositions
substantially better than the LSTM, ~7 times the accuracy of predicting
structured representations. By providing a concrete method to quantify
generalization for unseen combinations, we argue that structured
representations and compositional splits are a useful benchmark for image
captioning, and advocate compositional models that capture linguistic and
visual structure.
| [
{
"version": "v1",
"created": "Sat, 27 Aug 2016 00:34:00 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Atzmon",
"Yuval",
""
],
[
"Berant",
"Jonathan",
""
],
[
"Kezami",
"Vahid",
""
],
[
"Globerson",
"Amir",
""
],
[
"Chechik",
"Gal",
""
]
] | TITLE: Learning to generalize to new compositions in image understanding
ABSTRACT: Recurrent neural networks have recently been used for learning to describe
images using natural language. However, it has been observed that these models
generalize poorly to scenes that were not observed during training, possibly
depending too strongly on the statistics of the text in the training data. Here
we propose to describe images using short structured representations, aiming to
capture the crux of a description. These structured representations allow us to
tease-out and evaluate separately two types of generalization: standard
generalization to new images with similar scenes, and generalization to new
combinations of known entities. We compare two learning approaches on the
MS-COCO dataset: a state-of-the-art recurrent network based on an LSTM (Show,
Attend and Tell), and a simple structured prediction model on top of a deep
network. We find that the structured model generalizes to new compositions
substantially better than the LSTM, ~7 times the accuracy of predicting
structured representations. By providing a concrete method to quantify
generalization for unseen combinations, we argue that structured
representations and compositional splits are a useful benchmark for image
captioning, and advocate compositional models that capture linguistic and
visual structure.
| no_new_dataset | 0.946892 |
1608.07664 | Tian Lan | Jianhong Wang, Tian Lan, Xu Zhang, Limin Luo | Spatio-temporal Aware Non-negative Component Representation for Action
Recognition | 11 pages, 5 figures, 6 tables | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents a novel mid-level representation for action recognition,
named spatio-temporal aware non-negative component representation (STANNCR).
The proposed STANNCR is based on action component and incorporates the
spatial-temporal information. We first introduce a spatial-temporal
distribution vector (STDV) to model the distributions of local feature
locations in a compact and discriminative manner. Then we employ non-negative
matrix factorization (NMF) to learn the action components and encode the video
samples. The action component considers the correlations of visual words, which
effectively bridge the sematic gap in action recognition. To incorporate the
spatial-temporal cues for final representation, the STDV is used as the part of
graph regularization for NMF. The fusion of spatial-temporal information makes
the STANNCR more discriminative, and our fusion manner is more compact than
traditional method of concatenating vectors. The proposed approach is
extensively evaluated on three public datasets. The experimental results
demonstrate the effectiveness of STANNCR for action recognition.
| [
{
"version": "v1",
"created": "Sat, 27 Aug 2016 06:30:34 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Wang",
"Jianhong",
""
],
[
"Lan",
"Tian",
""
],
[
"Zhang",
"Xu",
""
],
[
"Luo",
"Limin",
""
]
] | TITLE: Spatio-temporal Aware Non-negative Component Representation for Action
Recognition
ABSTRACT: This paper presents a novel mid-level representation for action recognition,
named spatio-temporal aware non-negative component representation (STANNCR).
The proposed STANNCR is based on action component and incorporates the
spatial-temporal information. We first introduce a spatial-temporal
distribution vector (STDV) to model the distributions of local feature
locations in a compact and discriminative manner. Then we employ non-negative
matrix factorization (NMF) to learn the action components and encode the video
samples. The action component considers the correlations of visual words, which
effectively bridge the sematic gap in action recognition. To incorporate the
spatial-temporal cues for final representation, the STDV is used as the part of
graph regularization for NMF. The fusion of spatial-temporal information makes
the STANNCR more discriminative, and our fusion manner is more compact than
traditional method of concatenating vectors. The proposed approach is
extensively evaluated on three public datasets. The experimental results
demonstrate the effectiveness of STANNCR for action recognition.
| no_new_dataset | 0.945399 |
1608.07720 | Fei Li | Fei Li, Meishan Zhang, Guohong Fu, Tao Qian, Donghong Ji | A Bi-LSTM-RNN Model for Relation Classification Using Low-Cost Sequence
Features | null | null | null | null | cs.CL | http://creativecommons.org/publicdomain/zero/1.0/ | Relation classification is associated with many potential applications in the
artificial intelligence area. Recent approaches usually leverage neural
networks based on structure features such as syntactic or dependency features
to solve this problem. However, high-cost structure features make such
approaches inconvenient to be directly used. In addition, structure features
are probably domain-dependent. Therefore, this paper proposes a bi-directional
long-short-term-memory recurrent-neural-network (Bi-LSTM-RNN) model based on
low-cost sequence features to address relation classification. This model
divides a sentence or text segment into five parts, namely two target entities
and their three contexts. It learns the representations of entities and their
contexts, and uses them to classify relations. We evaluate our model on two
standard benchmark datasets in different domains, namely SemEval-2010 Task 8
and BioNLP-ST 2016 Task BB3. In the former dataset, our model achieves
comparable performance compared with other models using sequence features. In
the latter dataset, our model obtains the third best results compared with
other models in the official evaluation. Moreover, we find that the context
between two target entities plays the most important role in relation
classification. Furthermore, statistic experiments show that the context
between two target entities can be used as an approximate replacement of the
shortest dependency path when dependency parsing is not used.
| [
{
"version": "v1",
"created": "Sat, 27 Aug 2016 15:41:22 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Li",
"Fei",
""
],
[
"Zhang",
"Meishan",
""
],
[
"Fu",
"Guohong",
""
],
[
"Qian",
"Tao",
""
],
[
"Ji",
"Donghong",
""
]
] | TITLE: A Bi-LSTM-RNN Model for Relation Classification Using Low-Cost Sequence
Features
ABSTRACT: Relation classification is associated with many potential applications in the
artificial intelligence area. Recent approaches usually leverage neural
networks based on structure features such as syntactic or dependency features
to solve this problem. However, high-cost structure features make such
approaches inconvenient to be directly used. In addition, structure features
are probably domain-dependent. Therefore, this paper proposes a bi-directional
long-short-term-memory recurrent-neural-network (Bi-LSTM-RNN) model based on
low-cost sequence features to address relation classification. This model
divides a sentence or text segment into five parts, namely two target entities
and their three contexts. It learns the representations of entities and their
contexts, and uses them to classify relations. We evaluate our model on two
standard benchmark datasets in different domains, namely SemEval-2010 Task 8
and BioNLP-ST 2016 Task BB3. In the former dataset, our model achieves
comparable performance compared with other models using sequence features. In
the latter dataset, our model obtains the third best results compared with
other models in the official evaluation. Moreover, we find that the context
between two target entities plays the most important role in relation
classification. Furthermore, statistic experiments show that the context
between two target entities can be used as an approximate replacement of the
shortest dependency path when dependency parsing is not used.
| no_new_dataset | 0.952042 |
1608.07807 | Chandrajit M | Chandrajit M, Girisha R, Vasudev T and Ashok C B | Cast and Self Shadow Segmentation in Video Sequences using Interval
based Eigen Value Representation | 6 pages journal article | International Journal of Computer Applications 142(4):27-32, May
2016 | 10.5120/ijca2016909752 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Tracking of motion objects in the surveillance videos is useful for the
monitoring and analysis. The performance of the surveillance system will
deteriorate when shadows are detected as moving objects. Therefore, shadow
detection and elimination usually benefits the next stages. To overcome this
issue, a method for detection and elimination of shadows is proposed. This
paper presents a method for segmenting moving objects in video sequences based
on determining the Euclidian distance between two pixels considering
neighborhood values in temporal domain. Further, a method that segments cast
and self shadows in video sequences by computing the Eigen values for the
neighborhood of each pixel is proposed. The dual-map for cast and self shadow
pixels is represented based on the interval of Eigen values. The proposed
methods are tested on the benchmark IEEE CHANGE DETECTION 2014 dataset.
| [
{
"version": "v1",
"created": "Sun, 28 Aug 2016 13:07:16 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"M",
"Chandrajit",
""
],
[
"R",
"Girisha",
""
],
[
"T",
"Vasudev",
""
],
[
"B",
"Ashok C",
""
]
] | TITLE: Cast and Self Shadow Segmentation in Video Sequences using Interval
based Eigen Value Representation
ABSTRACT: Tracking of motion objects in the surveillance videos is useful for the
monitoring and analysis. The performance of the surveillance system will
deteriorate when shadows are detected as moving objects. Therefore, shadow
detection and elimination usually benefits the next stages. To overcome this
issue, a method for detection and elimination of shadows is proposed. This
paper presents a method for segmenting moving objects in video sequences based
on determining the Euclidian distance between two pixels considering
neighborhood values in temporal domain. Further, a method that segments cast
and self shadows in video sequences by computing the Eigen values for the
neighborhood of each pixel is proposed. The dual-map for cast and self shadow
pixels is represented based on the interval of Eigen values. The proposed
methods are tested on the benchmark IEEE CHANGE DETECTION 2014 dataset.
| no_new_dataset | 0.951863 |
1608.07916 | Bo Li | Bo Li, Tianlei Zhang, Tian Xia | Vehicle Detection from 3D Lidar Using Fully Convolutional Network | Robotics: Science and Systems, 2016 | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Convolutional network techniques have recently achieved great success in
vision based detection tasks. This paper introduces the recent development of
our research on transplanting the fully convolutional network technique to the
detection tasks on 3D range scan data. Specifically, the scenario is set as the
vehicle detection task from the range data of Velodyne 64E lidar. We proposes
to present the data in a 2D point map and use a single 2D end-to-end fully
convolutional network to predict the objectness confidence and the bounding
boxes simultaneously. By carefully design the bounding box encoding, it is able
to predict full 3D bounding boxes even using a 2D convolutional network.
Experiments on the KITTI dataset shows the state-of-the-art performance of the
proposed method.
| [
{
"version": "v1",
"created": "Mon, 29 Aug 2016 05:57:36 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Li",
"Bo",
""
],
[
"Zhang",
"Tianlei",
""
],
[
"Xia",
"Tian",
""
]
] | TITLE: Vehicle Detection from 3D Lidar Using Fully Convolutional Network
ABSTRACT: Convolutional network techniques have recently achieved great success in
vision based detection tasks. This paper introduces the recent development of
our research on transplanting the fully convolutional network technique to the
detection tasks on 3D range scan data. Specifically, the scenario is set as the
vehicle detection task from the range data of Velodyne 64E lidar. We proposes
to present the data in a 2D point map and use a single 2D end-to-end fully
convolutional network to predict the objectness confidence and the bounding
boxes simultaneously. By carefully design the bounding box encoding, it is able
to predict full 3D bounding boxes even using a 2D convolutional network.
Experiments on the KITTI dataset shows the state-of-the-art performance of the
proposed method.
| no_new_dataset | 0.948965 |
1608.07934 | Hadi Zare | Hadi Zare and Mojtaba Niazi | Relevant based structure learning for feature selection | 29 pages, 11 figures | Eng. Appl. Artif. Intel. 55 (2016) 93-102 | 10.1016/j.engappai.2016.06.001 | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Feature selection is an important task in many problems occurring in pattern
recognition, bioinformatics, machine learning and data mining applications. The
feature selection approach enables us to reduce the computation burden and the
falling accuracy effect of dealing with huge number of features in typical
learning problems. There is a variety of techniques for feature selection in
supervised learning problems based on different selection metrics. In this
paper, we propose a novel unified framework for feature selection built on the
graphical models and information theoretic tools. The proposed approach
exploits the structure learning among features to select more relevant and less
redundant features to the predictive modeling problem according to a primary
novel likelihood based criterion. In line with the selection of the optimal
subset of features through the proposed method, it provides us the Bayesian
network classifier without the additional cost of model training on the
selected subset of features. The optimal properties of our method are
established through empirical studies and computational complexity analysis.
Furthermore the proposed approach is evaluated on a bunch of benchmark datasets
based on the well-known classification algorithms. Extensive experiments
confirm the significant improvement of the proposed approach compared to the
earlier works.
| [
{
"version": "v1",
"created": "Mon, 29 Aug 2016 07:21:20 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Zare",
"Hadi",
""
],
[
"Niazi",
"Mojtaba",
""
]
] | TITLE: Relevant based structure learning for feature selection
ABSTRACT: Feature selection is an important task in many problems occurring in pattern
recognition, bioinformatics, machine learning and data mining applications. The
feature selection approach enables us to reduce the computation burden and the
falling accuracy effect of dealing with huge number of features in typical
learning problems. There is a variety of techniques for feature selection in
supervised learning problems based on different selection metrics. In this
paper, we propose a novel unified framework for feature selection built on the
graphical models and information theoretic tools. The proposed approach
exploits the structure learning among features to select more relevant and less
redundant features to the predictive modeling problem according to a primary
novel likelihood based criterion. In line with the selection of the optimal
subset of features through the proposed method, it provides us the Bayesian
network classifier without the additional cost of model training on the
selected subset of features. The optimal properties of our method are
established through empirical studies and computational complexity analysis.
Furthermore the proposed approach is evaluated on a bunch of benchmark datasets
based on the well-known classification algorithms. Extensive experiments
confirm the significant improvement of the proposed approach compared to the
earlier works.
| no_new_dataset | 0.948632 |
1608.07951 | Seoung Wug Oh | Seoung Wug Oh and Seon Joo Kim | Approaching the Computational Color Constancy as a Classification
Problem through Deep Learning | This is a preprint of an article accepted for publication in Pattern
Recognition, ELSEVIER | Pattern Recognition, Volume 61, January 2017, Pages 405 to 416 | 10.1016/j.patcog.2016.08.013 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Computational color constancy refers to the problem of computing the
illuminant color so that the images of a scene under varying illumination can
be normalized to an image under the canonical illumination. In this paper, we
adopt a deep learning framework for the illumination estimation problem. The
proposed method works under the assumption of uniform illumination over the
scene and aims for the accurate illuminant color computation. Specifically, we
trained the convolutional neural network to solve the problem by casting the
color constancy problem as an illumination classification problem. We designed
the deep learning architecture so that the output of the network can be
directly used for computing the color of the illumination. Experimental results
show that our deep network is able to extract useful features for the
illumination estimation and our method outperforms all previous color constancy
methods on multiple test datasets.
| [
{
"version": "v1",
"created": "Mon, 29 Aug 2016 08:41:55 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Oh",
"Seoung Wug",
""
],
[
"Kim",
"Seon Joo",
""
]
] | TITLE: Approaching the Computational Color Constancy as a Classification
Problem through Deep Learning
ABSTRACT: Computational color constancy refers to the problem of computing the
illuminant color so that the images of a scene under varying illumination can
be normalized to an image under the canonical illumination. In this paper, we
adopt a deep learning framework for the illumination estimation problem. The
proposed method works under the assumption of uniform illumination over the
scene and aims for the accurate illuminant color computation. Specifically, we
trained the convolutional neural network to solve the problem by casting the
color constancy problem as an illumination classification problem. We designed
the deep learning architecture so that the output of the network can be
directly used for computing the color of the illumination. Experimental results
show that our deep network is able to extract useful features for the
illumination estimation and our method outperforms all previous color constancy
methods on multiple test datasets.
| no_new_dataset | 0.948202 |
1608.08130 | Olaf Hartig | Magnus Knuth and Olaf Hartig and Harald Sack | Scheduling Refresh Queries for Keeping Results from a SPARQL Endpoint
Up-to-Date (Extended Version) | This document is an extended version of a paper published in ODBASE
2016 | null | null | null | cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Many datasets change over time. As a consequence, long-running applications
that cache and repeatedly use query results obtained from a SPARQL endpoint may
resubmit the queries regularly to ensure up-to-dateness of the results. While
this approach may be feasible if the number of such regular refresh queries is
manageable, with an increasing number of applications adopting this approach,
the SPARQL endpoint may become overloaded with such refresh queries. A more
scalable approach would be to use a middle-ware component at which the
applications register their queries and get notified with updated query results
once the results have changed. Then, this middle-ware can schedule the repeated
execution of the refresh queries without overloading the endpoint. In this
paper, we study the problem of scheduling refresh queries for a large number of
registered queries by assuming an overload-avoiding upper bound on the length
of a regular time slot available for testing refresh queries. We investigate a
variety of scheduling strategies and compare them experimentally in terms of
time slots needed before they recognize changes and number of changes that they
miss.
| [
{
"version": "v1",
"created": "Mon, 29 Aug 2016 16:16:36 GMT"
}
] | 2016-08-30T00:00:00 | [
[
"Knuth",
"Magnus",
""
],
[
"Hartig",
"Olaf",
""
],
[
"Sack",
"Harald",
""
]
] | TITLE: Scheduling Refresh Queries for Keeping Results from a SPARQL Endpoint
Up-to-Date (Extended Version)
ABSTRACT: Many datasets change over time. As a consequence, long-running applications
that cache and repeatedly use query results obtained from a SPARQL endpoint may
resubmit the queries regularly to ensure up-to-dateness of the results. While
this approach may be feasible if the number of such regular refresh queries is
manageable, with an increasing number of applications adopting this approach,
the SPARQL endpoint may become overloaded with such refresh queries. A more
scalable approach would be to use a middle-ware component at which the
applications register their queries and get notified with updated query results
once the results have changed. Then, this middle-ware can schedule the repeated
execution of the refresh queries without overloading the endpoint. In this
paper, we study the problem of scheduling refresh queries for a large number of
registered queries by assuming an overload-avoiding upper bound on the length
of a regular time slot available for testing refresh queries. We investigate a
variety of scheduling strategies and compare them experimentally in terms of
time slots needed before they recognize changes and number of changes that they
miss.
| no_new_dataset | 0.940079 |
1512.07071 | Lutz Bornmann Dr. | Lutz Bornmann, Robin Haunschild, Werner Marx | Policy documents as sources for measuring societal impact: How often is
climate change research mentioned in policy-related documents? | in press at Scientometrics | null | null | null | physics.soc-ph cs.DL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the current UK Research Excellence Framework (REF) and the Excellence in
Research for Australia (ERA) societal impact measurements are inherent parts of
the national evaluation systems. In this study, we deal with a relatively new
form of societal impact measurements. Recently, Altmetric - a start-up
providing publication level metrics - started to make data for publications
available which have been mentioned in policy documents. We regard this data
source as an interesting possibility to specifically measure the (societal)
impact of research. Using a comprehensive dataset with publications on climate
change as an example, we study the usefulness of the new data source for impact
measurement. Only 1.2% (n=2,341) out of 191,276 publications on climate change
in the dataset have at least one policy mention. We further reveal that papers
published in Nature and Science as well as from the areas "Earth and related
environmental sciences" and "Social and economic geography" are especially
relevant in the policy context. Given the low coverage of the climate change
literature in policy documents, this study can be only a first attempt to study
this new source of altmetric data. Further empirical studies are necessary in
upcoming years, because mentions in policy documents are of special interest in
the use of altmetric data for measuring target-oriented the broader impact of
research.
| [
{
"version": "v1",
"created": "Tue, 22 Dec 2015 13:09:52 GMT"
},
{
"version": "v2",
"created": "Fri, 26 Aug 2016 13:04:34 GMT"
}
] | 2016-08-29T00:00:00 | [
[
"Bornmann",
"Lutz",
""
],
[
"Haunschild",
"Robin",
""
],
[
"Marx",
"Werner",
""
]
] | TITLE: Policy documents as sources for measuring societal impact: How often is
climate change research mentioned in policy-related documents?
ABSTRACT: In the current UK Research Excellence Framework (REF) and the Excellence in
Research for Australia (ERA) societal impact measurements are inherent parts of
the national evaluation systems. In this study, we deal with a relatively new
form of societal impact measurements. Recently, Altmetric - a start-up
providing publication level metrics - started to make data for publications
available which have been mentioned in policy documents. We regard this data
source as an interesting possibility to specifically measure the (societal)
impact of research. Using a comprehensive dataset with publications on climate
change as an example, we study the usefulness of the new data source for impact
measurement. Only 1.2% (n=2,341) out of 191,276 publications on climate change
in the dataset have at least one policy mention. We further reveal that papers
published in Nature and Science as well as from the areas "Earth and related
environmental sciences" and "Social and economic geography" are especially
relevant in the policy context. Given the low coverage of the climate change
literature in policy documents, this study can be only a first attempt to study
this new source of altmetric data. Further empirical studies are necessary in
upcoming years, because mentions in policy documents are of special interest in
the use of altmetric data for measuring target-oriented the broader impact of
research.
| no_new_dataset | 0.932576 |
1602.06023 | Ramesh Nallapati | Ramesh Nallapati, Bowen Zhou, Cicero Nogueira dos santos, Caglar
Gulcehre, Bing Xiang | Abstractive Text Summarization Using Sequence-to-Sequence RNNs and
Beyond | null | The SIGNLL Conference on Computational Natural Language Learning
(CoNLL), 2016 | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we model abstractive text summarization using Attentional
Encoder-Decoder Recurrent Neural Networks, and show that they achieve
state-of-the-art performance on two different corpora. We propose several novel
models that address critical problems in summarization that are not adequately
modeled by the basic architecture, such as modeling key-words, capturing the
hierarchy of sentence-to-word structure, and emitting words that are rare or
unseen at training time. Our work shows that many of our proposed models
contribute to further improvement in performance. We also propose a new dataset
consisting of multi-sentence summaries, and establish performance benchmarks
for further research.
| [
{
"version": "v1",
"created": "Fri, 19 Feb 2016 02:04:18 GMT"
},
{
"version": "v2",
"created": "Mon, 11 Apr 2016 22:50:03 GMT"
},
{
"version": "v3",
"created": "Sat, 23 Apr 2016 02:38:01 GMT"
},
{
"version": "v4",
"created": "Wed, 10 Aug 2016 22:56:10 GMT"
},
{
"version": "v5",
"created": "Fri, 26 Aug 2016 16:13:13 GMT"
}
] | 2016-08-29T00:00:00 | [
[
"Nallapati",
"Ramesh",
""
],
[
"Zhou",
"Bowen",
""
],
[
"santos",
"Cicero Nogueira dos",
""
],
[
"Gulcehre",
"Caglar",
""
],
[
"Xiang",
"Bing",
""
]
] | TITLE: Abstractive Text Summarization Using Sequence-to-Sequence RNNs and
Beyond
ABSTRACT: In this work, we model abstractive text summarization using Attentional
Encoder-Decoder Recurrent Neural Networks, and show that they achieve
state-of-the-art performance on two different corpora. We propose several novel
models that address critical problems in summarization that are not adequately
modeled by the basic architecture, such as modeling key-words, capturing the
hierarchy of sentence-to-word structure, and emitting words that are rare or
unseen at training time. Our work shows that many of our proposed models
contribute to further improvement in performance. We also propose a new dataset
consisting of multi-sentence summaries, and establish performance benchmarks
for further research.
| new_dataset | 0.955569 |
1608.06010 | Yun Wang | Yun Wang, Xu Chen and Peter J. Ramadge | Feedback-Controlled Sequential Lasso Screening | null | null | null | null | cs.LG cs.AI cs.CV stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | One way to solve lasso problems when the dictionary does not fit into
available memory is to first screen the dictionary to remove unneeded features.
Prior research has shown that sequential screening methods offer the greatest
promise in this endeavor. Most existing work on sequential screening targets
the context of tuning parameter selection, where one screens and solves a
sequence of $N$ lasso problems with a fixed grid of geometrically spaced
regularization parameters. In contrast, we focus on the scenario where a target
regularization parameter has already been chosen via cross-validated model
selection, and we then need to solve many lasso instances using this fixed
value. In this context, we propose and explore a feedback controlled sequential
screening scheme. Feedback is used at each iteration to select the next problem
to be solved. This allows the sequence of problems to be adapted to the
instance presented and the number of intermediate problems to be automatically
selected. We demonstrate our feedback scheme using several datasets including a
dictionary of approximate size 100,000 by 300,000.
| [
{
"version": "v1",
"created": "Sun, 21 Aug 2016 23:40:56 GMT"
},
{
"version": "v2",
"created": "Thu, 25 Aug 2016 22:52:30 GMT"
}
] | 2016-08-29T00:00:00 | [
[
"Wang",
"Yun",
""
],
[
"Chen",
"Xu",
""
],
[
"Ramadge",
"Peter J.",
""
]
] | TITLE: Feedback-Controlled Sequential Lasso Screening
ABSTRACT: One way to solve lasso problems when the dictionary does not fit into
available memory is to first screen the dictionary to remove unneeded features.
Prior research has shown that sequential screening methods offer the greatest
promise in this endeavor. Most existing work on sequential screening targets
the context of tuning parameter selection, where one screens and solves a
sequence of $N$ lasso problems with a fixed grid of geometrically spaced
regularization parameters. In contrast, we focus on the scenario where a target
regularization parameter has already been chosen via cross-validated model
selection, and we then need to solve many lasso instances using this fixed
value. In this context, we propose and explore a feedback controlled sequential
screening scheme. Feedback is used at each iteration to select the next problem
to be solved. This allows the sequence of problems to be adapted to the
instance presented and the number of intermediate problems to be automatically
selected. We demonstrate our feedback scheme using several datasets including a
dictionary of approximate size 100,000 by 300,000.
| no_new_dataset | 0.946448 |
1608.06014 | Yun Wang | Yun Wang and Peter J. Ramadge | The Symmetry of a Simple Optimization Problem in Lasso Screening | null | null | null | null | cs.LG cs.AI cs.CV stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently dictionary screening has been proposed as an effective way to
improve the computational efficiency of solving the lasso problem, which is one
of the most commonly used method for learning sparse representations. To
address today's ever increasing large dataset, effective screening relies on a
tight region bound on the solution to the dual lasso. Typical region bounds are
in the form of an intersection of a sphere and multiple half spaces. One way to
tighten the region bound is using more half spaces, which however, adds to the
overhead of solving the high dimensional optimization problem in lasso
screening. This paper reveals the interesting property that the optimization
problem only depends on the projection of features onto the subspace spanned by
the normals of the half spaces. This property converts an optimization problem
in high dimension to much lower dimension, and thus sheds light on reducing the
computation overhead of lasso screening based on tighter region bounds.
| [
{
"version": "v1",
"created": "Sun, 21 Aug 2016 23:48:43 GMT"
},
{
"version": "v2",
"created": "Thu, 25 Aug 2016 22:05:24 GMT"
}
] | 2016-08-29T00:00:00 | [
[
"Wang",
"Yun",
""
],
[
"Ramadge",
"Peter J.",
""
]
] | TITLE: The Symmetry of a Simple Optimization Problem in Lasso Screening
ABSTRACT: Recently dictionary screening has been proposed as an effective way to
improve the computational efficiency of solving the lasso problem, which is one
of the most commonly used method for learning sparse representations. To
address today's ever increasing large dataset, effective screening relies on a
tight region bound on the solution to the dual lasso. Typical region bounds are
in the form of an intersection of a sphere and multiple half spaces. One way to
tighten the region bound is using more half spaces, which however, adds to the
overhead of solving the high dimensional optimization problem in lasso
screening. This paper reveals the interesting property that the optimization
problem only depends on the projection of features onto the subspace spanned by
the normals of the half spaces. This property converts an optimization problem
in high dimension to much lower dimension, and thus sheds light on reducing the
computation overhead of lasso screening based on tighter region bounds.
| no_new_dataset | 0.952175 |
1608.07411 | Wadim Kehl | Wadim Kehl, Tobias Holl, Federico Tombari, Slobodan Ilic, Nassir Navab | An Octree-Based Approach towards Efficient Variational Range Data Fusion | BMVC 2016 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Volume-based reconstruction is usually expensive both in terms of memory
consumption and runtime. Especially for sparse geometric structures, volumetric
representations produce a huge computational overhead. We present an efficient
way to fuse range data via a variational Octree-based minimization approach by
taking the actual range data geometry into account. We transform the data into
Octree-based truncated signed distance fields and show how the optimization can
be conducted on the newly created structures. The main challenge is to uphold
speed and a low memory footprint without sacrificing the solutions' accuracy
during optimization. We explain how to dynamically adjust the optimizer's
geometric structure via joining/splitting of Octree nodes and how to define the
operators. We evaluate on various datasets and outline the suitability in terms
of performance and geometric accuracy.
| [
{
"version": "v1",
"created": "Fri, 26 Aug 2016 10:01:51 GMT"
}
] | 2016-08-29T00:00:00 | [
[
"Kehl",
"Wadim",
""
],
[
"Holl",
"Tobias",
""
],
[
"Tombari",
"Federico",
""
],
[
"Ilic",
"Slobodan",
""
],
[
"Navab",
"Nassir",
""
]
] | TITLE: An Octree-Based Approach towards Efficient Variational Range Data Fusion
ABSTRACT: Volume-based reconstruction is usually expensive both in terms of memory
consumption and runtime. Especially for sparse geometric structures, volumetric
representations produce a huge computational overhead. We present an efficient
way to fuse range data via a variational Octree-based minimization approach by
taking the actual range data geometry into account. We transform the data into
Octree-based truncated signed distance fields and show how the optimization can
be conducted on the newly created structures. The main challenge is to uphold
speed and a low memory footprint without sacrificing the solutions' accuracy
during optimization. We explain how to dynamically adjust the optimizer's
geometric structure via joining/splitting of Octree nodes and how to define the
operators. We evaluate on various datasets and outline the suitability in terms
of performance and geometric accuracy.
| no_new_dataset | 0.950365 |
1608.07441 | Maxime Bucher | Maxime Bucher (Palaiseau), St\'ephane Herbin (Palaiseau), Fr\'ed\'eric
Jurie | Hard Negative Mining for Metric Learning Based Zero-Shot Classification | null | ECCV 16 WS TASK-CV: Transferring and Adapting Source Knowledge in
Computer Vision, Oct 2016, Amsterdam, Netherlands. ECCV 16 WS TASK-CV:
Transferring and Adapting Source Knowledge in Computer Vision | null | null | cs.LG cs.AI cs.CV stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Zero-Shot learning has been shown to be an efficient strategy for domain
adaptation. In this context, this paper builds on the recent work of Bucher et
al. [1], which proposed an approach to solve Zero-Shot classification problems
(ZSC) by introducing a novel metric learning based objective function. This
objective function allows to learn an optimal embedding of the attributes
jointly with a measure of similarity between images and attributes. This paper
extends their approach by proposing several schemes to control the generation
of the negative pairs, resulting in a significant improvement of the
performance and giving above state-of-the-art results on three challenging ZSC
datasets.
| [
{
"version": "v1",
"created": "Fri, 26 Aug 2016 12:42:43 GMT"
}
] | 2016-08-29T00:00:00 | [
[
"Bucher",
"Maxime",
"",
"Palaiseau"
],
[
"Herbin",
"Stéphane",
"",
"Palaiseau"
],
[
"Jurie",
"Frédéric",
""
]
] | TITLE: Hard Negative Mining for Metric Learning Based Zero-Shot Classification
ABSTRACT: Zero-Shot learning has been shown to be an efficient strategy for domain
adaptation. In this context, this paper builds on the recent work of Bucher et
al. [1], which proposed an approach to solve Zero-Shot classification problems
(ZSC) by introducing a novel metric learning based objective function. This
objective function allows to learn an optimal embedding of the attributes
jointly with a measure of similarity between images and attributes. This paper
extends their approach by proposing several schemes to control the generation
of the negative pairs, resulting in a significant improvement of the
performance and giving above state-of-the-art results on three challenging ZSC
datasets.
| no_new_dataset | 0.951639 |
1608.07454 | Vincent Lepetit | Tadej Vodopivec, Vincent Lepetit, Peter Peer | Fine Hand Segmentation using Convolutional Neural Networks | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a method for extracting very accurate masks of hands in egocentric
views. Our method is based on a novel Deep Learning architecture: In contrast
with current Deep Learning methods, we do not use upscaling layers applied to a
low-dimensional representation of the input image. Instead, we extract features
with convolutional layers and map them directly to a segmentation mask with a
fully connected layer. We show that this approach, when applied in a
multi-scale fashion, is both accurate and efficient enough for real-time. We
demonstrate it on a new dataset made of images captured in various
environments, from the outdoors to offices.
| [
{
"version": "v1",
"created": "Fri, 26 Aug 2016 13:40:08 GMT"
}
] | 2016-08-29T00:00:00 | [
[
"Vodopivec",
"Tadej",
""
],
[
"Lepetit",
"Vincent",
""
],
[
"Peer",
"Peter",
""
]
] | TITLE: Fine Hand Segmentation using Convolutional Neural Networks
ABSTRACT: We propose a method for extracting very accurate masks of hands in egocentric
views. Our method is based on a novel Deep Learning architecture: In contrast
with current Deep Learning methods, we do not use upscaling layers applied to a
low-dimensional representation of the input image. Instead, we extract features
with convolutional layers and map them directly to a segmentation mask with a
fully connected layer. We show that this approach, when applied in a
multi-scale fashion, is both accurate and efficient enough for real-time. We
demonstrate it on a new dataset made of images captured in various
environments, from the outdoors to offices.
| new_dataset | 0.950041 |
1510.01098 | Boshra Rajaei | Boshra Rajaei, Eric W. Tramel, Sylvain Gigan, Florent Krzakala,
Laurent Daudet | Intensity-only optical compressive imaging using a multiply scattering
material and a double phase retrieval approach | null | Proceedings of the 2016 IEEE International Conference on
Acoustics, Speech and Signal Processing (ICASSP) pages: 4054 - 4058 | 10.1109/ICASSP.2016.7472439 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, the problem of compressive imaging is addressed using natural
randomization by means of a multiply scattering medium. To utilize the medium
in this way, its corresponding transmission matrix must be estimated. To
calibrate the imager, we use a digital micromirror device (DMD) as a simple,
cheap, and high-resolution binary intensity modulator. We propose a phase
retrieval algorithm which is well adapted to intensity-only measurements on the
camera, and to the input binary intensity patterns, both to estimate the
complex transmission matrix as well as image reconstruction. We demonstrate
promising experimental results for the proposed algorithm using the MNIST
dataset of handwritten digits as example images.
| [
{
"version": "v1",
"created": "Mon, 5 Oct 2015 11:07:30 GMT"
},
{
"version": "v2",
"created": "Mon, 25 Jan 2016 14:35:44 GMT"
}
] | 2016-08-26T00:00:00 | [
[
"Rajaei",
"Boshra",
""
],
[
"Tramel",
"Eric W.",
""
],
[
"Gigan",
"Sylvain",
""
],
[
"Krzakala",
"Florent",
""
],
[
"Daudet",
"Laurent",
""
]
] | TITLE: Intensity-only optical compressive imaging using a multiply scattering
material and a double phase retrieval approach
ABSTRACT: In this paper, the problem of compressive imaging is addressed using natural
randomization by means of a multiply scattering medium. To utilize the medium
in this way, its corresponding transmission matrix must be estimated. To
calibrate the imager, we use a digital micromirror device (DMD) as a simple,
cheap, and high-resolution binary intensity modulator. We propose a phase
retrieval algorithm which is well adapted to intensity-only measurements on the
camera, and to the input binary intensity patterns, both to estimate the
complex transmission matrix as well as image reconstruction. We demonstrate
promising experimental results for the proposed algorithm using the MNIST
dataset of handwritten digits as example images.
| no_new_dataset | 0.952442 |
1603.07445 | Michael (Micky) Fire | Michael Fire and Carlos Guestrin | Time Is of the Essence: Analyzing the Effect of Vertex-Joining Time on
Complex Network Evolution | null | null | null | null | cs.SI physics.soc-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Complex networks have non-trivial characteristics and appear in many
real-world systems. Such networks are vitally important, but their full
underlying dynamics are not completely understood. Utilizing new data sources,
however, can unveil the evolution process of these networks.
This study uses the recently published Reddit dataset, containing over 1.65
billion comments, to construct the largest publicly available social network
corpus to date. We used this dataset to deeply examine the network evolution
process, which resulted in two key observations: First, links are more likely
to be created among users who join a network at a similar time. Second, the
rate in which new users join a network is a central factor in molding a
network's topology; i.e., different user-join patterns create different
topological properties.
Based on these observations, we developed the \textit{Temporal Preferential
Attachment} random network generation model. This model produces not only
scale-free networks that have relative high clustering coefficients, but also
networks that are sensitive to both the rate and the time in which users join
the network. This results in a more accurate and flexible model of how complex
networks evolve, one which more closely represents real-world data.
| [
{
"version": "v1",
"created": "Thu, 24 Mar 2016 05:32:31 GMT"
},
{
"version": "v2",
"created": "Thu, 25 Aug 2016 05:48:36 GMT"
}
] | 2016-08-26T00:00:00 | [
[
"Fire",
"Michael",
""
],
[
"Guestrin",
"Carlos",
""
]
] | TITLE: Time Is of the Essence: Analyzing the Effect of Vertex-Joining Time on
Complex Network Evolution
ABSTRACT: Complex networks have non-trivial characteristics and appear in many
real-world systems. Such networks are vitally important, but their full
underlying dynamics are not completely understood. Utilizing new data sources,
however, can unveil the evolution process of these networks.
This study uses the recently published Reddit dataset, containing over 1.65
billion comments, to construct the largest publicly available social network
corpus to date. We used this dataset to deeply examine the network evolution
process, which resulted in two key observations: First, links are more likely
to be created among users who join a network at a similar time. Second, the
rate in which new users join a network is a central factor in molding a
network's topology; i.e., different user-join patterns create different
topological properties.
Based on these observations, we developed the \textit{Temporal Preferential
Attachment} random network generation model. This model produces not only
scale-free networks that have relative high clustering coefficients, but also
networks that are sensitive to both the rate and the time in which users join
the network. This results in a more accurate and flexible model of how complex
networks evolve, one which more closely represents real-world data.
| no_new_dataset | 0.921428 |
1608.06985 | Ting-Chun Wang | Ting-Chun Wang, Jun-Yan Zhu, Ebi Hiroaki, Manmohan Chandraker, Alexei
A. Efros, Ravi Ramamoorthi | A 4D Light-Field Dataset and CNN Architectures for Material Recognition | European Conference on Computer Vision (ECCV) 2016 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a new light-field dataset of materials, and take advantage of
the recent success of deep learning to perform material recognition on the 4D
light-field. Our dataset contains 12 material categories, each with 100 images
taken with a Lytro Illum, from which we extract about 30,000 patches in total.
To the best of our knowledge, this is the first mid-size dataset for
light-field images. Our main goal is to investigate whether the additional
information in a light-field (such as multiple sub-aperture views and
view-dependent reflectance effects) can aid material recognition. Since
recognition networks have not been trained on 4D images before, we propose and
compare several novel CNN architectures to train on light-field images. In our
experiments, the best performing CNN architecture achieves a 7% boost compared
with 2D image classification (70% to 77%). These results constitute important
baselines that can spur further research in the use of CNNs for light-field
applications. Upon publication, our dataset also enables other novel
applications of light-fields, including object detection, image segmentation
and view interpolation.
| [
{
"version": "v1",
"created": "Wed, 24 Aug 2016 23:30:22 GMT"
}
] | 2016-08-26T00:00:00 | [
[
"Wang",
"Ting-Chun",
""
],
[
"Zhu",
"Jun-Yan",
""
],
[
"Hiroaki",
"Ebi",
""
],
[
"Chandraker",
"Manmohan",
""
],
[
"Efros",
"Alexei A.",
""
],
[
"Ramamoorthi",
"Ravi",
""
]
] | TITLE: A 4D Light-Field Dataset and CNN Architectures for Material Recognition
ABSTRACT: We introduce a new light-field dataset of materials, and take advantage of
the recent success of deep learning to perform material recognition on the 4D
light-field. Our dataset contains 12 material categories, each with 100 images
taken with a Lytro Illum, from which we extract about 30,000 patches in total.
To the best of our knowledge, this is the first mid-size dataset for
light-field images. Our main goal is to investigate whether the additional
information in a light-field (such as multiple sub-aperture views and
view-dependent reflectance effects) can aid material recognition. Since
recognition networks have not been trained on 4D images before, we propose and
compare several novel CNN architectures to train on light-field images. In our
experiments, the best performing CNN architecture achieves a 7% boost compared
with 2D image classification (70% to 77%). These results constitute important
baselines that can spur further research in the use of CNNs for light-field
applications. Upon publication, our dataset also enables other novel
applications of light-fields, including object detection, image segmentation
and view interpolation.
| new_dataset | 0.960805 |
1608.07138 | Cesar Roberto de Souza | C\'esar Roberto de Souza, Adrien Gaidon, Eleonora Vig, Antonio Manuel
L\'opez | Sympathy for the Details: Dense Trajectories and Hybrid Classification
Architectures for Action Recognition | Accepted for publication in the 14th European Conference on Computer
Vision (ECCV), Amsterdam, 2016, plus supplementary material | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Action recognition in videos is a challenging task due to the complexity of
the spatio-temporal patterns to model and the difficulty to acquire and learn
on large quantities of video data. Deep learning, although a breakthrough for
image classification and showing promise for videos, has still not clearly
superseded action recognition methods using hand-crafted features, even when
training on massive datasets. In this paper, we introduce hybrid video
classification architectures based on carefully designed unsupervised
representations of hand-crafted spatio-temporal features classified by
supervised deep networks. As we show in our experiments on five popular
benchmarks for action recognition, our hybrid model combines the best of both
worlds: it is data efficient (trained on 150 to 10000 short clips) and yet
improves significantly on the state of the art, including recent deep models
trained on millions of manually labelled images and videos.
| [
{
"version": "v1",
"created": "Thu, 25 Aug 2016 13:37:15 GMT"
}
] | 2016-08-26T00:00:00 | [
[
"de Souza",
"César Roberto",
""
],
[
"Gaidon",
"Adrien",
""
],
[
"Vig",
"Eleonora",
""
],
[
"López",
"Antonio Manuel",
""
]
] | TITLE: Sympathy for the Details: Dense Trajectories and Hybrid Classification
Architectures for Action Recognition
ABSTRACT: Action recognition in videos is a challenging task due to the complexity of
the spatio-temporal patterns to model and the difficulty to acquire and learn
on large quantities of video data. Deep learning, although a breakthrough for
image classification and showing promise for videos, has still not clearly
superseded action recognition methods using hand-crafted features, even when
training on massive datasets. In this paper, we introduce hybrid video
classification architectures based on carefully designed unsupervised
representations of hand-crafted spatio-temporal features classified by
supervised deep networks. As we show in our experiments on five popular
benchmarks for action recognition, our hybrid model combines the best of both
worlds: it is data efficient (trained on 150 to 10000 short clips) and yet
improves significantly on the state of the art, including recent deep models
trained on millions of manually labelled images and videos.
| no_new_dataset | 0.945096 |
1608.07242 | Mooyeol Baek | Hyeonseob Nam, Mooyeol Baek, Bohyung Han | Modeling and Propagating CNNs in a Tree Structure for Visual Tracking | 10 pages, Hyeonseob Nam and Mooyeol Baek have equal contribution | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present an online visual tracking algorithm by managing multiple target
appearance models in a tree structure. The proposed algorithm employs
Convolutional Neural Networks (CNNs) to represent target appearances, where
multiple CNNs collaborate to estimate target states and determine the desirable
paths for online model updates in the tree. By maintaining multiple CNNs in
diverse branches of tree structure, it is convenient to deal with
multi-modality in target appearances and preserve model reliability through
smooth updates along tree paths. Since multiple CNNs share all parameters in
convolutional layers, it takes advantage of multiple models with little extra
cost by saving memory space and avoiding redundant network evaluations. The
final target state is estimated by sampling target candidates around the state
in the previous frame and identifying the best sample in terms of a weighted
average score from a set of active CNNs. Our algorithm illustrates outstanding
performance compared to the state-of-the-art techniques in challenging datasets
such as online tracking benchmark and visual object tracking challenge.
| [
{
"version": "v1",
"created": "Thu, 25 Aug 2016 18:29:53 GMT"
}
] | 2016-08-26T00:00:00 | [
[
"Nam",
"Hyeonseob",
""
],
[
"Baek",
"Mooyeol",
""
],
[
"Han",
"Bohyung",
""
]
] | TITLE: Modeling and Propagating CNNs in a Tree Structure for Visual Tracking
ABSTRACT: We present an online visual tracking algorithm by managing multiple target
appearance models in a tree structure. The proposed algorithm employs
Convolutional Neural Networks (CNNs) to represent target appearances, where
multiple CNNs collaborate to estimate target states and determine the desirable
paths for online model updates in the tree. By maintaining multiple CNNs in
diverse branches of tree structure, it is convenient to deal with
multi-modality in target appearances and preserve model reliability through
smooth updates along tree paths. Since multiple CNNs share all parameters in
convolutional layers, it takes advantage of multiple models with little extra
cost by saving memory space and avoiding redundant network evaluations. The
final target state is estimated by sampling target candidates around the state
in the previous frame and identifying the best sample in terms of a weighted
average score from a set of active CNNs. Our algorithm illustrates outstanding
performance compared to the state-of-the-art techniques in challenging datasets
such as online tracking benchmark and visual object tracking challenge.
| no_new_dataset | 0.950319 |
1608.06757 | Sebastian Arnold | Sebastian Arnold, Felix A. Gers, Torsten Kilias, Alexander L\"oser | Robust Named Entity Recognition in Idiosyncratic Domains | 8 pages, 1 figure | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Named entity recognition often fails in idiosyncratic domains. That causes a
problem for depending tasks, such as entity linking and relation extraction. We
propose a generic and robust approach for high-recall named entity recognition.
Our approach is easy to train and offers strong generalization over diverse
domain-specific language, such as news documents (e.g. Reuters) or biomedical
text (e.g. Medline). Our approach is based on deep contextual sequence learning
and utilizes stacked bidirectional LSTM networks. Our model is trained with
only few hundred labeled sentences and does not rely on further external
knowledge. We report from our results F1 scores in the range of 84-94% on
standard datasets.
| [
{
"version": "v1",
"created": "Wed, 24 Aug 2016 09:06:14 GMT"
}
] | 2016-08-25T00:00:00 | [
[
"Arnold",
"Sebastian",
""
],
[
"Gers",
"Felix A.",
""
],
[
"Kilias",
"Torsten",
""
],
[
"Löser",
"Alexander",
""
]
] | TITLE: Robust Named Entity Recognition in Idiosyncratic Domains
ABSTRACT: Named entity recognition often fails in idiosyncratic domains. That causes a
problem for depending tasks, such as entity linking and relation extraction. We
propose a generic and robust approach for high-recall named entity recognition.
Our approach is easy to train and offers strong generalization over diverse
domain-specific language, such as news documents (e.g. Reuters) or biomedical
text (e.g. Medline). Our approach is based on deep contextual sequence learning
and utilizes stacked bidirectional LSTM networks. Our model is trained with
only few hundred labeled sentences and does not rely on further external
knowledge. We report from our results F1 scores in the range of 84-94% on
standard datasets.
| no_new_dataset | 0.95388 |
1608.06761 | Geetanjali Kale | Geetanjali Vinayak Kale and Varsha Hemant Patil | A Study of Vision based Human Motion Recognition and Analysis | 5 Figures, 18 Pages, International Journal of Ambient Computing and
Intelligence, Volume 7 Issue 2,July-December 2016 | null | 10.4018/IJACI.2016070104 | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Vision based human motion recognition has fascinated many researchers due to
its critical challenges and a variety of applications. The applications range
from simple gesture recognition to complicated behaviour understanding in
surveillance system. This leads to major development in the techniques related
to human motion representation and recognition. This paper discusses
applications, general framework of human motion recognition, and the details of
each of its components. The paper emphasizes on human motion representation and
the recognition methods along with their advantages and disadvantages. This
study also discusses the selected literature, popular datasets, and concludes
with the challenges in the domain along with a future direction. The human
motion recognition domain has been active for more than two decades, and has
provided a large amount of literature. A bird's eye view for new researchers in
the domain is presented in the paper.
| [
{
"version": "v1",
"created": "Wed, 24 Aug 2016 09:23:14 GMT"
}
] | 2016-08-25T00:00:00 | [
[
"Kale",
"Geetanjali Vinayak",
""
],
[
"Patil",
"Varsha Hemant",
""
]
] | TITLE: A Study of Vision based Human Motion Recognition and Analysis
ABSTRACT: Vision based human motion recognition has fascinated many researchers due to
its critical challenges and a variety of applications. The applications range
from simple gesture recognition to complicated behaviour understanding in
surveillance system. This leads to major development in the techniques related
to human motion representation and recognition. This paper discusses
applications, general framework of human motion recognition, and the details of
each of its components. The paper emphasizes on human motion representation and
the recognition methods along with their advantages and disadvantages. This
study also discusses the selected literature, popular datasets, and concludes
with the challenges in the domain along with a future direction. The human
motion recognition domain has been active for more than two decades, and has
provided a large amount of literature. A bird's eye view for new researchers in
the domain is presented in the paper.
| no_new_dataset | 0.950088 |
1608.06800 | Javier Alejandro Aldana Iuit | Javier Aldana-Iuit, Dmytro Mishkin, Ondrej Chum, Jiri Matas | In the Saddle: Chasing Fast and Repeatable Features | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A novel similarity-covariant feature detector that extracts points whose
neighbourhoods, when treated as a 3D intensity surface, have a saddle-like
intensity profile. The saddle condition is verified efficiently by intensity
comparisons on two concentric rings that must have exactly two dark-to-bright
and two bright-to-dark transitions satisfying certain geometric constraints.
Experiments show that the Saddle features are general, evenly spread and
appearing in high density in a range of images. The Saddle detector is among
the fastest proposed. In comparison with detector with similar speed, the
Saddle features show superior matching performance on number of challenging
datasets.
| [
{
"version": "v1",
"created": "Wed, 24 Aug 2016 12:57:34 GMT"
}
] | 2016-08-25T00:00:00 | [
[
"Aldana-Iuit",
"Javier",
""
],
[
"Mishkin",
"Dmytro",
""
],
[
"Chum",
"Ondrej",
""
],
[
"Matas",
"Jiri",
""
]
] | TITLE: In the Saddle: Chasing Fast and Repeatable Features
ABSTRACT: A novel similarity-covariant feature detector that extracts points whose
neighbourhoods, when treated as a 3D intensity surface, have a saddle-like
intensity profile. The saddle condition is verified efficiently by intensity
comparisons on two concentric rings that must have exactly two dark-to-bright
and two bright-to-dark transitions satisfying certain geometric constraints.
Experiments show that the Saddle features are general, evenly spread and
appearing in high density in a range of images. The Saddle detector is among
the fastest proposed. In comparison with detector with similar speed, the
Saddle features show superior matching performance on number of challenging
datasets.
| no_new_dataset | 0.920647 |
1608.06861 | Jun Yue | Xia Yue, Wang Man, Jun Yue, Guangcao Liu | Parallel K-Medoids++ Spatial Clustering Algorithm Based on MapReduce | null | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Clustering analysis has received considerable attention in spatial data
mining for several years. With the rapid development of the geospatial
information technologies, the size of spatial information data is growing
exponentially which makes clustering massive spatial data a challenging task.
In order to improve the efficiency of spatial clustering for large scale data,
many researchers proposed several efficient clustering algorithms in parallel.
In this paper, a new K-Medoids++ spatial clustering algorithm based on
MapReduce for clustering massive spatial data is proposed. The initialization
algorithm to decrease the number of iterations is combined with the MapReduce
framework. Comparative Experiments conducted over different dataset and
different number of nodes indicate that the proposed K-Medoids spatial
clustering algorithm provides better efficiency than traditional K-Medoids and
scales well while processing massive spatial data on commodity hardware.
| [
{
"version": "v1",
"created": "Wed, 24 Aug 2016 15:21:24 GMT"
}
] | 2016-08-25T00:00:00 | [
[
"Yue",
"Xia",
""
],
[
"Man",
"Wang",
""
],
[
"Yue",
"Jun",
""
],
[
"Liu",
"Guangcao",
""
]
] | TITLE: Parallel K-Medoids++ Spatial Clustering Algorithm Based on MapReduce
ABSTRACT: Clustering analysis has received considerable attention in spatial data
mining for several years. With the rapid development of the geospatial
information technologies, the size of spatial information data is growing
exponentially which makes clustering massive spatial data a challenging task.
In order to improve the efficiency of spatial clustering for large scale data,
many researchers proposed several efficient clustering algorithms in parallel.
In this paper, a new K-Medoids++ spatial clustering algorithm based on
MapReduce for clustering massive spatial data is proposed. The initialization
algorithm to decrease the number of iterations is combined with the MapReduce
framework. Comparative Experiments conducted over different dataset and
different number of nodes indicate that the proposed K-Medoids spatial
clustering algorithm provides better efficiency than traditional K-Medoids and
scales well while processing massive spatial data on commodity hardware.
| no_new_dataset | 0.955152 |
1608.06863 | Victoria Peterson Mrs | Victoria Peterson, Hugo Leonardo Rufiner, Ruben Daniel Spies | Kullback-Leibler Penalized Sparse Discriminant Analysis for
Event-Related Potential Classification | 27 pages, 4 figures | null | null | null | cs.CV cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A brain computer interface (BCI) is a system which provides direct
communication between the mind of a person and the outside world by using only
brain activity (EEG). The event-related potential (ERP)-based BCI problem
consists of a binary pattern recognition. Linear discriminant analysis (LDA) is
widely used to solve this type of classification problems, but it fails when
the number of features is large relative to the number of observations. In this
work we propose a penalized version of the sparse discriminant analysis (SDA),
called Kullback-Leibler penalized sparse discriminant analysis (KLSDA). This
method inherits both the discriminative feature selection and classification
properties of SDA and it also improves SDA performance through the addition of
Kullback-Leibler class discrepancy information. The KLSDA method is design to
automatically select the optimal regularization parameters. Numerical
experiments with two real ERP-EEG datasets show that this new method
outperforms standard SDA.
| [
{
"version": "v1",
"created": "Wed, 24 Aug 2016 15:32:51 GMT"
}
] | 2016-08-25T00:00:00 | [
[
"Peterson",
"Victoria",
""
],
[
"Rufiner",
"Hugo Leonardo",
""
],
[
"Spies",
"Ruben Daniel",
""
]
] | TITLE: Kullback-Leibler Penalized Sparse Discriminant Analysis for
Event-Related Potential Classification
ABSTRACT: A brain computer interface (BCI) is a system which provides direct
communication between the mind of a person and the outside world by using only
brain activity (EEG). The event-related potential (ERP)-based BCI problem
consists of a binary pattern recognition. Linear discriminant analysis (LDA) is
widely used to solve this type of classification problems, but it fails when
the number of features is large relative to the number of observations. In this
work we propose a penalized version of the sparse discriminant analysis (SDA),
called Kullback-Leibler penalized sparse discriminant analysis (KLSDA). This
method inherits both the discriminative feature selection and classification
properties of SDA and it also improves SDA performance through the addition of
Kullback-Leibler class discrepancy information. The KLSDA method is design to
automatically select the optimal regularization parameters. Numerical
experiments with two real ERP-EEG datasets show that this new method
outperforms standard SDA.
| no_new_dataset | 0.94625 |
1505.07717 | Rodrigo Cabral Farias | Rodrigo Cabral Farias, Jeremy Emile Cohen and Pierre Comon | Exploring multimodal data fusion through joint decompositions with
flexible couplings | 15 pages, 7 figures, revised version | null | 10.1109/TSP.2016.2576425 | null | cs.IT math.IT | http://creativecommons.org/licenses/by-nc-sa/4.0/ | A Bayesian framework is proposed to define flexible coupling models for joint
tensor decompositions of multiple data sets. Under this framework, a natural
formulation of the data fusion problem is to cast it in terms of a joint
maximum a posteriori (MAP) estimator. Data driven scenarios of joint posterior
distributions are provided, including general Gaussian priors and non Gaussian
coupling priors. We present and discuss implementation issues of algorithms
used to obtain the joint MAP estimator. We also show how this framework can be
adapted to tackle the problem of joint decompositions of large datasets. In the
case of a conditional Gaussian coupling with a linear transformation, we give
theoretical bounds on the data fusion performance using the Bayesian Cramer-Rao
bound. Simulations are reported for hybrid coupling models ranging from simple
additive Gaussian models, to Gamma-type models with positive variables and to
the coupling of data sets which are inherently of different size due to
different resolution of the measurement devices.
| [
{
"version": "v1",
"created": "Thu, 28 May 2015 15:07:14 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Oct 2015 07:01:43 GMT"
},
{
"version": "v3",
"created": "Wed, 25 May 2016 12:03:01 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Farias",
"Rodrigo Cabral",
""
],
[
"Cohen",
"Jeremy Emile",
""
],
[
"Comon",
"Pierre",
""
]
] | TITLE: Exploring multimodal data fusion through joint decompositions with
flexible couplings
ABSTRACT: A Bayesian framework is proposed to define flexible coupling models for joint
tensor decompositions of multiple data sets. Under this framework, a natural
formulation of the data fusion problem is to cast it in terms of a joint
maximum a posteriori (MAP) estimator. Data driven scenarios of joint posterior
distributions are provided, including general Gaussian priors and non Gaussian
coupling priors. We present and discuss implementation issues of algorithms
used to obtain the joint MAP estimator. We also show how this framework can be
adapted to tackle the problem of joint decompositions of large datasets. In the
case of a conditional Gaussian coupling with a linear transformation, we give
theoretical bounds on the data fusion performance using the Bayesian Cramer-Rao
bound. Simulations are reported for hybrid coupling models ranging from simple
additive Gaussian models, to Gamma-type models with positive variables and to
the coupling of data sets which are inherently of different size due to
different resolution of the measurement devices.
| no_new_dataset | 0.94474 |
1508.00842 | Sougata Chaudhuri | Sougata Chaudhuri and Ambuj Tewari | Perceptron like Algorithms for Online Learning to Rank | Under review in Journal of Artificial Intelligence Research (JAIR) | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Perceptron is a classic online algorithm for learning a classification
function. In this paper, we provide a novel extension of the perceptron
algorithm to the learning to rank problem in information retrieval. We consider
popular listwise performance measures such as Normalized Discounted Cumulative
Gain (NDCG) and Average Precision (AP). A modern perspective on perceptron for
classification is that it is simply an instance of online gradient descent
(OGD), during mistake rounds, using the hinge loss function. Motivated by this
interpretation, we propose a novel family of listwise, large margin ranking
surrogates. Members of this family can be thought of as analogs of the hinge
loss. Exploiting a certain self-bounding property of the proposed family, we
provide a guarantee on the cumulative NDCG (or AP) induced loss incurred by our
perceptron-like algorithm. We show that, if there exists a perfect oracle
ranker which can correctly rank each instance in an online sequence of ranking
data, with some margin, the cumulative loss of perceptron algorithm on that
sequence is bounded by a constant, irrespective of the length of the sequence.
This result is reminiscent of Novikoff's convergence theorem for the
classification perceptron. Moreover, we prove a lower bound on the cumulative
loss achievable by any deterministic algorithm, under the assumption of
existence of perfect oracle ranker. The lower bound shows that our perceptron
bound is not tight, and we propose another, \emph{purely online}, algorithm
which achieves the lower bound. We provide empirical results on simulated and
large commercial datasets to corroborate our theoretical results.
| [
{
"version": "v1",
"created": "Tue, 4 Aug 2015 17:23:46 GMT"
},
{
"version": "v2",
"created": "Sun, 6 Mar 2016 20:32:04 GMT"
},
{
"version": "v3",
"created": "Sun, 27 Mar 2016 00:29:55 GMT"
},
{
"version": "v4",
"created": "Tue, 23 Aug 2016 06:52:04 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Chaudhuri",
"Sougata",
""
],
[
"Tewari",
"Ambuj",
""
]
] | TITLE: Perceptron like Algorithms for Online Learning to Rank
ABSTRACT: Perceptron is a classic online algorithm for learning a classification
function. In this paper, we provide a novel extension of the perceptron
algorithm to the learning to rank problem in information retrieval. We consider
popular listwise performance measures such as Normalized Discounted Cumulative
Gain (NDCG) and Average Precision (AP). A modern perspective on perceptron for
classification is that it is simply an instance of online gradient descent
(OGD), during mistake rounds, using the hinge loss function. Motivated by this
interpretation, we propose a novel family of listwise, large margin ranking
surrogates. Members of this family can be thought of as analogs of the hinge
loss. Exploiting a certain self-bounding property of the proposed family, we
provide a guarantee on the cumulative NDCG (or AP) induced loss incurred by our
perceptron-like algorithm. We show that, if there exists a perfect oracle
ranker which can correctly rank each instance in an online sequence of ranking
data, with some margin, the cumulative loss of perceptron algorithm on that
sequence is bounded by a constant, irrespective of the length of the sequence.
This result is reminiscent of Novikoff's convergence theorem for the
classification perceptron. Moreover, we prove a lower bound on the cumulative
loss achievable by any deterministic algorithm, under the assumption of
existence of perfect oracle ranker. The lower bound shows that our perceptron
bound is not tight, and we propose another, \emph{purely online}, algorithm
which achieves the lower bound. We provide empirical results on simulated and
large commercial datasets to corroborate our theoretical results.
| no_new_dataset | 0.942295 |
1511.04695 | Linxiao Yang | Linxiao Yang and Jun Fang and Hongbin Li and Bing Zeng | An Iterative Reweighted Method for Tucker Decomposition of Incomplete
Multiway Tensors | null | null | 10.1109/TSP.2016.2572047 | null | cs.NA cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider the problem of low-rank decomposition of incomplete multiway
tensors. Since many real-world data lie on an intrinsically low dimensional
subspace, tensor low-rank decomposition with missing entries has applications
in many data analysis problems such as recommender systems and image
inpainting. In this paper, we focus on Tucker decomposition which represents an
Nth-order tensor in terms of N factor matrices and a core tensor via
multilinear operations. To exploit the underlying multilinear low-rank
structure in high-dimensional datasets, we propose a group-based log-sum
penalty functional to place structural sparsity over the core tensor, which
leads to a compact representation with smallest core tensor. The method for
Tucker decomposition is developed by iteratively minimizing a surrogate
function that majorizes the original objective function, which results in an
iterative reweighted process. In addition, to reduce the computational
complexity, an over-relaxed monotone fast iterative shrinkage-thresholding
technique is adapted and embedded in the iterative reweighted process. The
proposed method is able to determine the model complexity (i.e. multilinear
rank) in an automatic way. Simulation results show that the proposed algorithm
offers competitive performance compared with other existing algorithms.
| [
{
"version": "v1",
"created": "Sun, 15 Nov 2015 12:56:36 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Yang",
"Linxiao",
""
],
[
"Fang",
"Jun",
""
],
[
"Li",
"Hongbin",
""
],
[
"Zeng",
"Bing",
""
]
] | TITLE: An Iterative Reweighted Method for Tucker Decomposition of Incomplete
Multiway Tensors
ABSTRACT: We consider the problem of low-rank decomposition of incomplete multiway
tensors. Since many real-world data lie on an intrinsically low dimensional
subspace, tensor low-rank decomposition with missing entries has applications
in many data analysis problems such as recommender systems and image
inpainting. In this paper, we focus on Tucker decomposition which represents an
Nth-order tensor in terms of N factor matrices and a core tensor via
multilinear operations. To exploit the underlying multilinear low-rank
structure in high-dimensional datasets, we propose a group-based log-sum
penalty functional to place structural sparsity over the core tensor, which
leads to a compact representation with smallest core tensor. The method for
Tucker decomposition is developed by iteratively minimizing a surrogate
function that majorizes the original objective function, which results in an
iterative reweighted process. In addition, to reduce the computational
complexity, an over-relaxed monotone fast iterative shrinkage-thresholding
technique is adapted and embedded in the iterative reweighted process. The
proposed method is able to determine the model complexity (i.e. multilinear
rank) in an automatic way. Simulation results show that the proposed algorithm
offers competitive performance compared with other existing algorithms.
| no_new_dataset | 0.944791 |
1603.00961 | Jan Egger | Tobias L\"uddemann, Jan Egger | Interactive and Scale Invariant Segmentation of the Rectum/Sigmoid via
User-Defined Templates | 6 pages, 4 figures, 1 table, 43 references | SPIE Medical Imaging Conference 2016, Paper 9784-113 | 10.1117/12.2216226 | null | cs.CV cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Among all types of cancer, gynecological malignancies belong to the 4th most
frequent type of cancer among women. Besides chemotherapy and external beam
radiation, brachytherapy is the standard procedure for the treatment of these
malignancies. In the progress of treatment planning, localization of the tumor
as the target volume and adjacent organs of risks by segmentation is crucial to
accomplish an optimal radiation distribution to the tumor while simultaneously
preserving healthy tissue. Segmentation is performed manually and represents a
time-consuming task in clinical daily routine. This study focuses on the
segmentation of the rectum/sigmoid colon as an Organ-At-Risk in gynecological
brachytherapy. The proposed segmentation method uses an interactive,
graph-based segmentation scheme with a user-defined template. The scheme
creates a directed two dimensional graph, followed by the minimal cost closed
set computation on the graph, resulting in an outlining of the rectum. The
graphs outline is dynamically adapted to the last calculated cut. Evaluation
was performed by comparing manual segmentations of the rectum/sigmoid colon to
results achieved with the proposed method. The comparison of the algorithmic to
manual results yielded to a Dice Similarity Coefficient value of 83.85+/-4.08%,
in comparison to 83.97+/-8.08% for the comparison of two manual segmentations
of the same physician. Utilizing the proposed methodology resulted in a median
time of 128 seconds per dataset, compared to 300 seconds needed for pure manual
segmentation.
| [
{
"version": "v1",
"created": "Thu, 3 Mar 2016 03:39:59 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Lüddemann",
"Tobias",
""
],
[
"Egger",
"Jan",
""
]
] | TITLE: Interactive and Scale Invariant Segmentation of the Rectum/Sigmoid via
User-Defined Templates
ABSTRACT: Among all types of cancer, gynecological malignancies belong to the 4th most
frequent type of cancer among women. Besides chemotherapy and external beam
radiation, brachytherapy is the standard procedure for the treatment of these
malignancies. In the progress of treatment planning, localization of the tumor
as the target volume and adjacent organs of risks by segmentation is crucial to
accomplish an optimal radiation distribution to the tumor while simultaneously
preserving healthy tissue. Segmentation is performed manually and represents a
time-consuming task in clinical daily routine. This study focuses on the
segmentation of the rectum/sigmoid colon as an Organ-At-Risk in gynecological
brachytherapy. The proposed segmentation method uses an interactive,
graph-based segmentation scheme with a user-defined template. The scheme
creates a directed two dimensional graph, followed by the minimal cost closed
set computation on the graph, resulting in an outlining of the rectum. The
graphs outline is dynamically adapted to the last calculated cut. Evaluation
was performed by comparing manual segmentations of the rectum/sigmoid colon to
results achieved with the proposed method. The comparison of the algorithmic to
manual results yielded to a Dice Similarity Coefficient value of 83.85+/-4.08%,
in comparison to 83.97+/-8.08% for the comparison of two manual segmentations
of the same physician. Utilizing the proposed methodology resulted in a median
time of 128 seconds per dataset, compared to 300 seconds needed for pure manual
segmentation.
| no_new_dataset | 0.953319 |
1603.01855 | Sougata Chaudhuri | Sougata Chaudhuri and Ambuj Tewari | Online Learning to Rank with Feedback at the Top | Appearing in AISTATS 2016 | AISTATS 16, volume 51 of JMLR Workshop and Conference Proceedings,
pg.-277-285, 2016 | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider an online learning to rank setting in which, at each round, an
oblivious adversary generates a list of $m$ documents, pertaining to a query,
and the learner produces scores to rank the documents. The adversary then
generates a relevance vector and the learner updates its ranker according to
the feedback received. We consider the setting where the feedback is restricted
to be the relevance levels of only the top $k$ documents in the ranked list for
$k \ll m$. However, the performance of learner is judged based on the
unrevealed full relevance vectors, using an appropriate learning to rank loss
function. We develop efficient algorithms for well known losses in the
pointwise, pairwise and listwise families. We also prove that no online
algorithm can have sublinear regret, with top-1 feedback, for any loss that is
calibrated with respect to NDCG. We apply our algorithms on benchmark datasets
demonstrating efficient online learning of a ranking function from highly
restricted feedback.
| [
{
"version": "v1",
"created": "Sun, 6 Mar 2016 18:43:54 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Chaudhuri",
"Sougata",
""
],
[
"Tewari",
"Ambuj",
""
]
] | TITLE: Online Learning to Rank with Feedback at the Top
ABSTRACT: We consider an online learning to rank setting in which, at each round, an
oblivious adversary generates a list of $m$ documents, pertaining to a query,
and the learner produces scores to rank the documents. The adversary then
generates a relevance vector and the learner updates its ranker according to
the feedback received. We consider the setting where the feedback is restricted
to be the relevance levels of only the top $k$ documents in the ranked list for
$k \ll m$. However, the performance of learner is judged based on the
unrevealed full relevance vectors, using an appropriate learning to rank loss
function. We develop efficient algorithms for well known losses in the
pointwise, pairwise and listwise families. We also prove that no online
algorithm can have sublinear regret, with top-1 feedback, for any loss that is
calibrated with respect to NDCG. We apply our algorithms on benchmark datasets
demonstrating efficient online learning of a ranking function from highly
restricted feedback.
| no_new_dataset | 0.942823 |
1608.05477 | Xi Peng | Xi Peng, Rogerio S. Feris, Xiaoyu Wang, Dimitris N. Metaxas | A Recurrent Encoder-Decoder Network for Sequential Face Alignment | European Conference on Computer Vision (ECCV), 2016 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a novel recurrent encoder-decoder network model for real-time
video-based face alignment. Our proposed model predicts 2D facial point maps
regularized by a regression loss, while uniquely exploiting recurrent learning
at both spatial and temporal dimensions. At the spatial level, we add a
feedback loop connection between the combined output response map and the
input, in order to enable iterative coarse-to-fine face alignment using a
single network model. At the temporal level, we first decouple the features in
the bottleneck of the network into temporal-variant factors, such as pose and
expression, and temporal-invariant factors, such as identity information.
Temporal recurrent learning is then applied to the decoupled temporal-variant
features, yielding better generalization and significantly more accurate
results at test time. We perform a comprehensive experimental analysis, showing
the importance of each component of our proposed model, as well as superior
results over the state-of-the-art in standard datasets.
| [
{
"version": "v1",
"created": "Fri, 19 Aug 2016 02:28:50 GMT"
},
{
"version": "v2",
"created": "Tue, 23 Aug 2016 01:23:48 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Peng",
"Xi",
""
],
[
"Feris",
"Rogerio S.",
""
],
[
"Wang",
"Xiaoyu",
""
],
[
"Metaxas",
"Dimitris N.",
""
]
] | TITLE: A Recurrent Encoder-Decoder Network for Sequential Face Alignment
ABSTRACT: We propose a novel recurrent encoder-decoder network model for real-time
video-based face alignment. Our proposed model predicts 2D facial point maps
regularized by a regression loss, while uniquely exploiting recurrent learning
at both spatial and temporal dimensions. At the spatial level, we add a
feedback loop connection between the combined output response map and the
input, in order to enable iterative coarse-to-fine face alignment using a
single network model. At the temporal level, we first decouple the features in
the bottleneck of the network into temporal-variant factors, such as pose and
expression, and temporal-invariant factors, such as identity information.
Temporal recurrent learning is then applied to the decoupled temporal-variant
features, yielding better generalization and significantly more accurate
results at test time. We perform a comprehensive experimental analysis, showing
the importance of each component of our proposed model, as well as superior
results over the state-of-the-art in standard datasets.
| no_new_dataset | 0.949482 |
1608.06169 | Jaroslaw Szlichta | Jaroslaw Szlichta, Parke Godfrey, Lukasz Golab, Mehdi Kargar, Divesh
Srivastava | Effective and Complete Discovery of Order Dependencies via Set-based
Axiomatization | 14 pages | null | null | null | cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Integrity constraints (ICs) provide a valuable tool for expressing and
enforcing application semantics. However, formulating constraints manually
requires domain expertise, is prone to human errors, and may be excessively
time consuming, especially on large datasets. Hence, proposals for automatic
discovery have been made for some classes of ICs, such as functional
dependencies (FDs), and recently, order dependencies (ODs). ODs properly
subsume FDs, as they can additionally express business rules involving order;
e.g., an employee never has a higher salary while paying lower taxes compared
with another employee.
We address the limitations of prior work on OD discovery which has factorial
complexity in the number of attributes, is incomplete (i.e., it does not
discover valid ODs that cannot be inferred from the ones found) and is not
concise (i.e., it can result in "redundant" discovery and overly large
discovery sets). We improve significantly on complexity, offer completeness,
and define a compact canonical form. This is based on a novel polynomial
mapping to a canonical form for ODs, and a sound and complete set of axioms
(inference rules) for canonical ODs. This allows us to develop an efficient
set-containment, lattice-driven OD discovery algorithm that uses the inference
rules to prune the search space. Our algorithm has exponential worst-case time
complexity in the number of attributes and linear complexity in the number of
tuples. We prove that it produces a complete, minimal set of ODs (i.e., minimal
with regards to the canonical representation). Finally, using real and
synthetic datasets, we experimentally show orders-of-magnitude performance
improvements over the current state-of-the-art algorithm and demonstrate
effectiveness of our techniques.
| [
{
"version": "v1",
"created": "Mon, 22 Aug 2016 14:03:46 GMT"
},
{
"version": "v2",
"created": "Tue, 23 Aug 2016 19:54:06 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Szlichta",
"Jaroslaw",
""
],
[
"Godfrey",
"Parke",
""
],
[
"Golab",
"Lukasz",
""
],
[
"Kargar",
"Mehdi",
""
],
[
"Srivastava",
"Divesh",
""
]
] | TITLE: Effective and Complete Discovery of Order Dependencies via Set-based
Axiomatization
ABSTRACT: Integrity constraints (ICs) provide a valuable tool for expressing and
enforcing application semantics. However, formulating constraints manually
requires domain expertise, is prone to human errors, and may be excessively
time consuming, especially on large datasets. Hence, proposals for automatic
discovery have been made for some classes of ICs, such as functional
dependencies (FDs), and recently, order dependencies (ODs). ODs properly
subsume FDs, as they can additionally express business rules involving order;
e.g., an employee never has a higher salary while paying lower taxes compared
with another employee.
We address the limitations of prior work on OD discovery which has factorial
complexity in the number of attributes, is incomplete (i.e., it does not
discover valid ODs that cannot be inferred from the ones found) and is not
concise (i.e., it can result in "redundant" discovery and overly large
discovery sets). We improve significantly on complexity, offer completeness,
and define a compact canonical form. This is based on a novel polynomial
mapping to a canonical form for ODs, and a sound and complete set of axioms
(inference rules) for canonical ODs. This allows us to develop an efficient
set-containment, lattice-driven OD discovery algorithm that uses the inference
rules to prune the search space. Our algorithm has exponential worst-case time
complexity in the number of attributes and linear complexity in the number of
tuples. We prove that it produces a complete, minimal set of ODs (i.e., minimal
with regards to the canonical representation). Finally, using real and
synthetic datasets, we experimentally show orders-of-magnitude performance
improvements over the current state-of-the-art algorithm and demonstrate
effectiveness of our techniques.
| no_new_dataset | 0.943971 |
1608.06298 | Jonathan Gemmell Jonathan Gemmell | Greg Zanotti, Miller Horvath, Lucas Nunes Barbosa, Venkata Trinadh
Kumar Gupta Immedisetty, Jonathan Gemmell | Infusing Collaborative Recommenders with Distributed Representations | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recommender systems assist users in navigating complex information spaces and
focus their attention on the content most relevant to their needs. Often these
systems rely on user activity or descriptions of the content. Social annotation
systems, in which users collaboratively assign tags to items, provide another
means to capture information about users and items. Each of these data sources
provides unique benefits, capturing different relationships.
In this paper, we propose leveraging multiple sources of data: ratings data
as users report their affinity toward an item, tagging data as users assign
annotations to items, and item data collected from an online database. Taken
together, these datasets provide the opportunity to learn rich distributed
representations by exploiting recent advances in neural network architectures.
We first produce representations that subjectively capture interesting
relationships among the data. We then empirically evaluate the utility of the
representations to predict a user's rating on an item and show that it
outperforms more traditional representations. Finally, we demonstrate that
traditional representations can be combined with representations trained
through a neural network to achieve even better results.
| [
{
"version": "v1",
"created": "Mon, 22 Aug 2016 20:05:01 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Zanotti",
"Greg",
""
],
[
"Horvath",
"Miller",
""
],
[
"Barbosa",
"Lucas Nunes",
""
],
[
"Immedisetty",
"Venkata Trinadh Kumar Gupta",
""
],
[
"Gemmell",
"Jonathan",
""
]
] | TITLE: Infusing Collaborative Recommenders with Distributed Representations
ABSTRACT: Recommender systems assist users in navigating complex information spaces and
focus their attention on the content most relevant to their needs. Often these
systems rely on user activity or descriptions of the content. Social annotation
systems, in which users collaboratively assign tags to items, provide another
means to capture information about users and items. Each of these data sources
provides unique benefits, capturing different relationships.
In this paper, we propose leveraging multiple sources of data: ratings data
as users report their affinity toward an item, tagging data as users assign
annotations to items, and item data collected from an online database. Taken
together, these datasets provide the opportunity to learn rich distributed
representations by exploiting recent advances in neural network architectures.
We first produce representations that subjectively capture interesting
relationships among the data. We then empirically evaluate the utility of the
representations to predict a user's rating on an item and show that it
outperforms more traditional representations. Finally, we demonstrate that
traditional representations can be combined with representations trained
through a neural network to achieve even better results.
| no_new_dataset | 0.943504 |
1608.06408 | Sougata Chaudhuri | Sougata Chaudhuri and Ambuj Tewari | Online Learning to Rank with Top-k Feedback | Under review in JMLR | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider two settings of online learning to rank where feedback is
restricted to top ranked items. The problem is cast as an online game between a
learner and sequence of users, over $T$ rounds. In both settings, the learners
objective is to present ranked list of items to the users. The learner's
performance is judged on the entire ranked list and true relevances of the
items. However, the learner receives highly restricted feedback at end of each
round, in form of relevances of only the top $k$ ranked items, where $k \ll m$.
The first setting is \emph{non-contextual}, where the list of items to be
ranked is fixed. The second setting is \emph{contextual}, where lists of items
vary, in form of traditional query-document lists. No stochastic assumption is
made on the generation process of relevances of items and contexts. We provide
efficient ranking strategies for both the settings. The strategies achieve
$O(T^{2/3})$ regret, where regret is based on popular ranking measures in first
setting and ranking surrogates in second setting. We also provide impossibility
results for certain ranking measures and a certain class of surrogates, when
feedback is restricted to the top ranked item, i.e. $k=1$. We empirically
demonstrate the performance of our algorithms on simulated and real world
datasets.
| [
{
"version": "v1",
"created": "Tue, 23 Aug 2016 07:40:08 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Chaudhuri",
"Sougata",
""
],
[
"Tewari",
"Ambuj",
""
]
] | TITLE: Online Learning to Rank with Top-k Feedback
ABSTRACT: We consider two settings of online learning to rank where feedback is
restricted to top ranked items. The problem is cast as an online game between a
learner and sequence of users, over $T$ rounds. In both settings, the learners
objective is to present ranked list of items to the users. The learner's
performance is judged on the entire ranked list and true relevances of the
items. However, the learner receives highly restricted feedback at end of each
round, in form of relevances of only the top $k$ ranked items, where $k \ll m$.
The first setting is \emph{non-contextual}, where the list of items to be
ranked is fixed. The second setting is \emph{contextual}, where lists of items
vary, in form of traditional query-document lists. No stochastic assumption is
made on the generation process of relevances of items and contexts. We provide
efficient ranking strategies for both the settings. The strategies achieve
$O(T^{2/3})$ regret, where regret is based on popular ranking measures in first
setting and ranking surrogates in second setting. We also provide impossibility
results for certain ranking measures and a certain class of surrogates, when
feedback is restricted to the top ranked item, i.e. $k=1$. We empirically
demonstrate the performance of our algorithms on simulated and real world
datasets.
| no_new_dataset | 0.950227 |
1608.06495 | Dan Xu | Nannan Li, Dan Xu, Zhenqiang Ying, Zhihao Li, Ge Li | Searching Action Proposals via Spatial Actionness Estimation and
Temporal Path Inference and Tracking | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we address the problem of searching action proposals in
unconstrained video clips. Our approach starts from actionness estimation on
frame-level bounding boxes, and then aggregates the bounding boxes belonging to
the same actor across frames via linking, associating, tracking to generate
spatial-temporal continuous action paths. To achieve the target, a novel
actionness estimation method is firstly proposed by utilizing both human
appearance and motion cues. Then, the association of the action paths is
formulated as a maximum set coverage problem with the results of actionness
estimation as a priori. To further promote the performance, we design an
improved optimization objective for the problem and provide a greedy search
algorithm to solve it. Finally, a tracking-by-detection scheme is designed to
further refine the searched action paths. Extensive experiments on two
challenging datasets, UCF-Sports and UCF-101, show that the proposed approach
advances state-of-the-art proposal generation performance in terms of both
accuracy and proposal quantity.
| [
{
"version": "v1",
"created": "Tue, 23 Aug 2016 13:08:30 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Li",
"Nannan",
""
],
[
"Xu",
"Dan",
""
],
[
"Ying",
"Zhenqiang",
""
],
[
"Li",
"Zhihao",
""
],
[
"Li",
"Ge",
""
]
] | TITLE: Searching Action Proposals via Spatial Actionness Estimation and
Temporal Path Inference and Tracking
ABSTRACT: In this paper, we address the problem of searching action proposals in
unconstrained video clips. Our approach starts from actionness estimation on
frame-level bounding boxes, and then aggregates the bounding boxes belonging to
the same actor across frames via linking, associating, tracking to generate
spatial-temporal continuous action paths. To achieve the target, a novel
actionness estimation method is firstly proposed by utilizing both human
appearance and motion cues. Then, the association of the action paths is
formulated as a maximum set coverage problem with the results of actionness
estimation as a priori. To further promote the performance, we design an
improved optimization objective for the problem and provide a greedy search
algorithm to solve it. Finally, a tracking-by-detection scheme is designed to
further refine the searched action paths. Extensive experiments on two
challenging datasets, UCF-Sports and UCF-101, show that the proposed approach
advances state-of-the-art proposal generation performance in terms of both
accuracy and proposal quantity.
| no_new_dataset | 0.945651 |
1608.06557 | Le Hou | Le Hou, Dimitris Samaras, Tahsin M. Kurc, Yi Gao, Joel H. Saltz | Neural Networks with Smooth Adaptive Activation Functions for Regression | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In Neural Networks (NN), Adaptive Activation Functions (AAF) have parameters
that control the shapes of activation functions. These parameters are trained
along with other parameters in the NN. AAFs have improved performance of Neural
Networks (NN) in multiple classification tasks. In this paper, we propose and
apply AAFs on feedforward NNs for regression tasks. We argue that applying AAFs
in the regression (second-to-last) layer of a NN can significantly decrease the
bias of the regression NN. However, using existing AAFs may lead to
overfitting. To address this problem, we propose a Smooth Adaptive Activation
Function (SAAF) with piecewise polynomial form which can approximate any
continuous function to arbitrary degree of error. NNs with SAAFs can avoid
overfitting by simply regularizing the parameters. In particular, an NN with
SAAFs is Lipschitz continuous given a bounded magnitude of the NN parameters.
We prove an upper-bound for model complexity in terms of fat-shattering
dimension for any Lipschitz continuous regression model. Thus, regularizing the
parameters in NNs with SAAFs avoids overfitting. We empirically evaluated NNs
with SAAFs and achieved state-of-the-art results on multiple regression
datasets.
| [
{
"version": "v1",
"created": "Tue, 23 Aug 2016 15:56:08 GMT"
}
] | 2016-08-24T00:00:00 | [
[
"Hou",
"Le",
""
],
[
"Samaras",
"Dimitris",
""
],
[
"Kurc",
"Tahsin M.",
""
],
[
"Gao",
"Yi",
""
],
[
"Saltz",
"Joel H.",
""
]
] | TITLE: Neural Networks with Smooth Adaptive Activation Functions for Regression
ABSTRACT: In Neural Networks (NN), Adaptive Activation Functions (AAF) have parameters
that control the shapes of activation functions. These parameters are trained
along with other parameters in the NN. AAFs have improved performance of Neural
Networks (NN) in multiple classification tasks. In this paper, we propose and
apply AAFs on feedforward NNs for regression tasks. We argue that applying AAFs
in the regression (second-to-last) layer of a NN can significantly decrease the
bias of the regression NN. However, using existing AAFs may lead to
overfitting. To address this problem, we propose a Smooth Adaptive Activation
Function (SAAF) with piecewise polynomial form which can approximate any
continuous function to arbitrary degree of error. NNs with SAAFs can avoid
overfitting by simply regularizing the parameters. In particular, an NN with
SAAFs is Lipschitz continuous given a bounded magnitude of the NN parameters.
We prove an upper-bound for model complexity in terms of fat-shattering
dimension for any Lipschitz continuous regression model. Thus, regularizing the
parameters in NNs with SAAFs avoids overfitting. We empirically evaluated NNs
with SAAFs and achieved state-of-the-art results on multiple regression
datasets.
| no_new_dataset | 0.950503 |
1405.4897 | Yun Wang | Zhen James Xiang, Yun Wang and Peter J. Ramadge | Screening Tests for Lasso Problems | Accepted to IEEE Transactions on Pattern Analysis and Machine
Intelligence | null | 10.1109/TPAMI.2016.2568185 | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper is a survey of dictionary screening for the lasso problem. The
lasso problem seeks a sparse linear combination of the columns of a dictionary
to best match a given target vector. This sparse representation has proven
useful in a variety of subsequent processing and decision tasks. For a given
target vector, dictionary screening quickly identifies a subset of dictionary
columns that will receive zero weight in a solution of the corresponding lasso
problem. These columns can be removed from the dictionary prior to solving the
lasso problem without impacting the optimality of the solution obtained. This
has two potential advantages: it reduces the size of the dictionary, allowing
the lasso problem to be solved with less resources, and it may speed up
obtaining a solution. Using a geometrically intuitive framework, we provide
basic insights for understanding useful lasso screening tests and their
limitations. We also provide illustrative numerical studies on several
datasets.
| [
{
"version": "v1",
"created": "Mon, 19 May 2014 21:07:08 GMT"
},
{
"version": "v2",
"created": "Sun, 21 Aug 2016 22:04:31 GMT"
}
] | 2016-08-23T00:00:00 | [
[
"Xiang",
"Zhen James",
""
],
[
"Wang",
"Yun",
""
],
[
"Ramadge",
"Peter J.",
""
]
] | TITLE: Screening Tests for Lasso Problems
ABSTRACT: This paper is a survey of dictionary screening for the lasso problem. The
lasso problem seeks a sparse linear combination of the columns of a dictionary
to best match a given target vector. This sparse representation has proven
useful in a variety of subsequent processing and decision tasks. For a given
target vector, dictionary screening quickly identifies a subset of dictionary
columns that will receive zero weight in a solution of the corresponding lasso
problem. These columns can be removed from the dictionary prior to solving the
lasso problem without impacting the optimality of the solution obtained. This
has two potential advantages: it reduces the size of the dictionary, allowing
the lasso problem to be solved with less resources, and it may speed up
obtaining a solution. Using a geometrically intuitive framework, we provide
basic insights for understanding useful lasso screening tests and their
limitations. We also provide illustrative numerical studies on several
datasets.
| no_new_dataset | 0.95275 |
1508.02848 | Yunjin Chen | Yunjin Chen and Thomas Pock | Trainable Nonlinear Reaction Diffusion: A Flexible Framework for Fast
and Effective Image Restoration | 14 pages, 13 figures, to appear in IEEE Transactions on Pattern
Analysis and Machine Intelligence (TPAMI) | null | 10.1109/TPAMI.2016.2596743 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Image restoration is a long-standing problem in low-level computer vision
with many interesting applications. We describe a flexible learning framework
based on the concept of nonlinear reaction diffusion models for various image
restoration problems. By embodying recent improvements in nonlinear diffusion
models, we propose a dynamic nonlinear reaction diffusion model with
time-dependent parameters (\ie, linear filters and influence functions). In
contrast to previous nonlinear diffusion models, all the parameters, including
the filters and the influence functions, are simultaneously learned from
training data through a loss based approach. We call this approach TNRD --
\textit{Trainable Nonlinear Reaction Diffusion}. The TNRD approach is
applicable for a variety of image restoration tasks by incorporating
appropriate reaction force. We demonstrate its capabilities with three
representative applications, Gaussian image denoising, single image super
resolution and JPEG deblocking. Experiments show that our trained nonlinear
diffusion models largely benefit from the training of the parameters and
finally lead to the best reported performance on common test datasets for the
tested applications. Our trained models preserve the structural simplicity of
diffusion models and take only a small number of diffusion steps, thus are
highly efficient. Moreover, they are also well-suited for parallel computation
on GPUs, which makes the inference procedure extremely fast.
| [
{
"version": "v1",
"created": "Wed, 12 Aug 2015 08:40:48 GMT"
},
{
"version": "v2",
"created": "Sat, 20 Aug 2016 04:48:54 GMT"
}
] | 2016-08-23T00:00:00 | [
[
"Chen",
"Yunjin",
""
],
[
"Pock",
"Thomas",
""
]
] | TITLE: Trainable Nonlinear Reaction Diffusion: A Flexible Framework for Fast
and Effective Image Restoration
ABSTRACT: Image restoration is a long-standing problem in low-level computer vision
with many interesting applications. We describe a flexible learning framework
based on the concept of nonlinear reaction diffusion models for various image
restoration problems. By embodying recent improvements in nonlinear diffusion
models, we propose a dynamic nonlinear reaction diffusion model with
time-dependent parameters (\ie, linear filters and influence functions). In
contrast to previous nonlinear diffusion models, all the parameters, including
the filters and the influence functions, are simultaneously learned from
training data through a loss based approach. We call this approach TNRD --
\textit{Trainable Nonlinear Reaction Diffusion}. The TNRD approach is
applicable for a variety of image restoration tasks by incorporating
appropriate reaction force. We demonstrate its capabilities with three
representative applications, Gaussian image denoising, single image super
resolution and JPEG deblocking. Experiments show that our trained nonlinear
diffusion models largely benefit from the training of the parameters and
finally lead to the best reported performance on common test datasets for the
tested applications. Our trained models preserve the structural simplicity of
diffusion models and take only a small number of diffusion steps, thus are
highly efficient. Moreover, they are also well-suited for parallel computation
on GPUs, which makes the inference procedure extremely fast.
| no_new_dataset | 0.949201 |
1511.00792 | Sayantan Dasgupta | Sayantan Dasgupta | Fast Collaborative Filtering from Implicit Feedback with Provable
Guarantees | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Building recommendation algorithms is one of the most challenging tasks in
Machine Learning. Although most of the recommendation systems are built on
explicit feedback available from the users in terms of rating or text, a
majority of the applications do not receive such feedback. Here we consider the
recommendation task where the only available data is the records of user-item
interaction over web applications over time, in terms of subscription or
purchase of items; this is known as implicit feedback recommendation. There is
usually a massive amount of such user-item interaction available for any web
applications. Algorithms like PLSI or Matrix Factorization runs several
iterations through the dataset, and may prove very expensive for large
datasets. Here we propose a recommendation algorithm based on Method of Moment,
which involves factorization of second and third order moments of the dataset.
Our algorithm can be proven to be globally convergent using PAC learning
theory. Further, we show how to extract the parameters using only three passes
through the entire dataset. This results in a highly scalable algorithm that
scales up to million of users even on a machine with a single-core processor
and 8 GB RAM and produces competitive performance in comparison with existing
algorithms.
| [
{
"version": "v1",
"created": "Tue, 3 Nov 2015 06:43:54 GMT"
},
{
"version": "v10",
"created": "Sun, 21 Aug 2016 08:00:15 GMT"
},
{
"version": "v2",
"created": "Wed, 4 Nov 2015 06:26:04 GMT"
},
{
"version": "v3",
"created": "Thu, 5 Nov 2015 13:06:07 GMT"
},
{
"version": "v4",
"created": "Tue, 10 Nov 2015 12:05:40 GMT"
},
{
"version": "v5",
"created": "Mon, 23 Nov 2015 10:05:40 GMT"
},
{
"version": "v6",
"created": "Thu, 24 Dec 2015 20:33:13 GMT"
},
{
"version": "v7",
"created": "Fri, 15 Jan 2016 15:44:30 GMT"
},
{
"version": "v8",
"created": "Sat, 6 Feb 2016 10:28:25 GMT"
},
{
"version": "v9",
"created": "Wed, 8 Jun 2016 16:33:38 GMT"
}
] | 2016-08-23T00:00:00 | [
[
"Dasgupta",
"Sayantan",
""
]
] | TITLE: Fast Collaborative Filtering from Implicit Feedback with Provable
Guarantees
ABSTRACT: Building recommendation algorithms is one of the most challenging tasks in
Machine Learning. Although most of the recommendation systems are built on
explicit feedback available from the users in terms of rating or text, a
majority of the applications do not receive such feedback. Here we consider the
recommendation task where the only available data is the records of user-item
interaction over web applications over time, in terms of subscription or
purchase of items; this is known as implicit feedback recommendation. There is
usually a massive amount of such user-item interaction available for any web
applications. Algorithms like PLSI or Matrix Factorization runs several
iterations through the dataset, and may prove very expensive for large
datasets. Here we propose a recommendation algorithm based on Method of Moment,
which involves factorization of second and third order moments of the dataset.
Our algorithm can be proven to be globally convergent using PAC learning
theory. Further, we show how to extract the parameters using only three passes
through the entire dataset. This results in a highly scalable algorithm that
scales up to million of users even on a machine with a single-core processor
and 8 GB RAM and produces competitive performance in comparison with existing
algorithms.
| no_new_dataset | 0.937897 |
1603.08148 | \c{C}a\u{g}lar G\"ul\c{c}ehre | Caglar Gulcehre, Sungjin Ahn, Ramesh Nallapati, Bowen Zhou and Yoshua
Bengio | Pointing the Unknown Words | ACL 2016 Oral Paper | null | null | null | cs.CL cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The problem of rare and unknown words is an important issue that can
potentially influence the performance of many NLP systems, including both the
traditional count-based and the deep learning models. We propose a novel way to
deal with the rare and unseen words for the neural network models using
attention. Our model uses two softmax layers in order to predict the next word
in conditional language models: one predicts the location of a word in the
source sentence, and the other predicts a word in the shortlist vocabulary. At
each time-step, the decision of which softmax layer to use choose adaptively
made by an MLP which is conditioned on the context.~We motivate our work from a
psychological evidence that humans naturally have a tendency to point towards
objects in the context or the environment when the name of an object is not
known.~We observe improvements on two tasks, neural machine translation on the
Europarl English to French parallel corpora and text summarization on the
Gigaword dataset using our proposed model.
| [
{
"version": "v1",
"created": "Sat, 26 Mar 2016 22:31:57 GMT"
},
{
"version": "v2",
"created": "Sun, 3 Apr 2016 21:12:57 GMT"
},
{
"version": "v3",
"created": "Sun, 21 Aug 2016 20:03:39 GMT"
}
] | 2016-08-23T00:00:00 | [
[
"Gulcehre",
"Caglar",
""
],
[
"Ahn",
"Sungjin",
""
],
[
"Nallapati",
"Ramesh",
""
],
[
"Zhou",
"Bowen",
""
],
[
"Bengio",
"Yoshua",
""
]
] | TITLE: Pointing the Unknown Words
ABSTRACT: The problem of rare and unknown words is an important issue that can
potentially influence the performance of many NLP systems, including both the
traditional count-based and the deep learning models. We propose a novel way to
deal with the rare and unseen words for the neural network models using
attention. Our model uses two softmax layers in order to predict the next word
in conditional language models: one predicts the location of a word in the
source sentence, and the other predicts a word in the shortlist vocabulary. At
each time-step, the decision of which softmax layer to use choose adaptively
made by an MLP which is conditioned on the context.~We motivate our work from a
psychological evidence that humans naturally have a tendency to point towards
objects in the context or the environment when the name of an object is not
known.~We observe improvements on two tasks, neural machine translation on the
Europarl English to French parallel corpora and text summarization on the
Gigaword dataset using our proposed model.
| no_new_dataset | 0.951051 |
1608.00276 | Jeroen Vuurens | Jeroen B. P. Vuurens, Martha Larson, Arjen P. de Vries | Exploring Deep Space: Learning Personalized Ranking in a Semantic Space | 6 pages, RecSys 2016 RSDL workshop | null | 10.1145/2988450.2988457 | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recommender systems leverage both content and user interactions to generate
recommendations that fit users' preferences. The recent surge of interest in
deep learning presents new opportunities for exploiting these two sources of
information. To recommend items we propose to first learn a user-independent
high-dimensional semantic space in which items are positioned according to
their substitutability, and then learn a user-specific transformation function
to transform this space into a ranking according to the user's past
preferences. An advantage of the proposed architecture is that it can be used
to effectively recommend items using either content that describes the items or
user-item ratings. We show that this approach significantly outperforms
state-of-the-art recommender systems on the MovieLens 1M dataset.
| [
{
"version": "v1",
"created": "Sun, 31 Jul 2016 22:38:46 GMT"
},
{
"version": "v2",
"created": "Mon, 22 Aug 2016 18:43:04 GMT"
}
] | 2016-08-23T00:00:00 | [
[
"Vuurens",
"Jeroen B. P.",
""
],
[
"Larson",
"Martha",
""
],
[
"de Vries",
"Arjen P.",
""
]
] | TITLE: Exploring Deep Space: Learning Personalized Ranking in a Semantic Space
ABSTRACT: Recommender systems leverage both content and user interactions to generate
recommendations that fit users' preferences. The recent surge of interest in
deep learning presents new opportunities for exploiting these two sources of
information. To recommend items we propose to first learn a user-independent
high-dimensional semantic space in which items are positioned according to
their substitutability, and then learn a user-specific transformation function
to transform this space into a ranking according to the user's past
preferences. An advantage of the proposed architecture is that it can be used
to effectively recommend items using either content that describes the items or
user-item ratings. We show that this approach significantly outperforms
state-of-the-art recommender systems on the MovieLens 1M dataset.
| no_new_dataset | 0.953362 |
1608.01413 | Subhro Roy | Subhro Roy and Dan Roth | Solving General Arithmetic Word Problems | EMNLP 2015 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | This paper presents a novel approach to automatically solving arithmetic word
problems. This is the first algorithmic approach that can handle arithmetic
problems with multiple steps and operations, without depending on additional
annotations or predefined templates. We develop a theory for expression trees
that can be used to represent and evaluate the target arithmetic expressions;
we use it to uniquely decompose the target arithmetic problem to multiple
classification problems; we then compose an expression tree, combining these
with world knowledge through a constrained inference framework. Our classifiers
gain from the use of {\em quantity schemas} that supports better extraction of
features. Experimental results show that our method outperforms existing
systems, achieving state of the art performance on benchmark datasets of
arithmetic word problems.
| [
{
"version": "v1",
"created": "Thu, 4 Aug 2016 01:47:23 GMT"
},
{
"version": "v2",
"created": "Sat, 20 Aug 2016 11:50:41 GMT"
}
] | 2016-08-23T00:00:00 | [
[
"Roy",
"Subhro",
""
],
[
"Roth",
"Dan",
""
]
] | TITLE: Solving General Arithmetic Word Problems
ABSTRACT: This paper presents a novel approach to automatically solving arithmetic word
problems. This is the first algorithmic approach that can handle arithmetic
problems with multiple steps and operations, without depending on additional
annotations or predefined templates. We develop a theory for expression trees
that can be used to represent and evaluate the target arithmetic expressions;
we use it to uniquely decompose the target arithmetic problem to multiple
classification problems; we then compose an expression tree, combining these
with world knowledge through a constrained inference framework. Our classifiers
gain from the use of {\em quantity schemas} that supports better extraction of
features. Experimental results show that our method outperforms existing
systems, achieving state of the art performance on benchmark datasets of
arithmetic word problems.
| no_new_dataset | 0.941975 |
1608.05777 | Lei Xu | Lei Xu, Ziyun Wang, Ayana, Zhiyuan Liu, Maosong Sun | Topic Sensitive Neural Headline Generation | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Neural models have recently been used in text summarization including
headline generation. The model can be trained using a set of document-headline
pairs. However, the model does not explicitly consider topical similarities and
differences of documents. We suggest to categorizing documents into various
topics so that documents within the same topic are similar in content and share
similar summarization patterns. Taking advantage of topic information of
documents, we propose topic sensitive neural headline generation model. Our
model can generate more accurate summaries guided by document topics. We test
our model on LCSTS dataset, and experiments show that our method outperforms
other baselines on each topic and achieves the state-of-art performance.
| [
{
"version": "v1",
"created": "Sat, 20 Aug 2016 03:43:29 GMT"
}
] | 2016-08-23T00:00:00 | [
[
"Xu",
"Lei",
""
],
[
"Wang",
"Ziyun",
""
],
[
"Ayana",
"",
""
],
[
"Liu",
"Zhiyuan",
""
],
[
"Sun",
"Maosong",
""
]
] | TITLE: Topic Sensitive Neural Headline Generation
ABSTRACT: Neural models have recently been used in text summarization including
headline generation. The model can be trained using a set of document-headline
pairs. However, the model does not explicitly consider topical similarities and
differences of documents. We suggest to categorizing documents into various
topics so that documents within the same topic are similar in content and share
similar summarization patterns. Taking advantage of topic information of
documents, we propose topic sensitive neural headline generation model. Our
model can generate more accurate summaries guided by document topics. We test
our model on LCSTS dataset, and experiments show that our method outperforms
other baselines on each topic and achieves the state-of-art performance.
| no_new_dataset | 0.949153 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.