S. Hossein Hosseini, Tohid Nouri, Afshin Ebrahimi, S. Ali Hosseini
Comments: 6 pages, 15 figures, 1 table, IEEE International Conference on Signal Processing and Intelligent System (ICSPIS16), Dec. 2016
Subjects: Optimization and Control (math.OC); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
We propose a swarm-based optimization algorithm inspired by air currents of a
tornado. Two main air currents – spiral and updraft – are mimicked. Spiral
motion is designed for exploration of new search areas and updraft movements is
deployed for exploitation of a promising candidate solution. Assignment of just
one search direction to each particle at each iteration, leads to low
computational complexity of the proposed algorithm respect to the conventional
algorithms. Regardless of the step size parameters, the only parameter of the
proposed algorithm, called tornado diameter, can be efficiently adjusted by
randomization. Numerical results over six different benchmark cost functions
indicate comparable and, in some cases, better performance of the proposed
algorithm respect to some other metaheuristics.
Zhiyuan Zha, Xinggan Zhang, Qiong Wang, Yechao Bai, Lan Tang
Comments: arXiv admin note: text overlap with arXiv:1609.03302
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Nonlocal image representation has been successfully used in many
image-related inverse problems including denoising, deblurring and deblocking.
However, a majority of reconstruction methods only exploit the nonlocal
self-similarity (NSS) prior of the degraded observation image, it is very
challenging to reconstruct the latent clean image. In this paper we propose a
novel model for image denoising via group sparsity residual and external NSS
prior. To boost the performance of image denoising, the concept of group
sparsity residual is proposed, and thus the problem of image denoising is
transformed into one that reduces the group sparsity residual. Due to the fact
that the groups contain a large amount of NSS information of natural images, we
obtain a good estimation of the group sparse coefficients of the original image
by the external NSS prior based on Gaussian Mixture model (GMM) learning and
the group sparse coefficients of noisy image is used to approximate the
estimation. Experimental results have demonstrated that the proposed method not
only outperforms many state-of-the-art methods, but also delivers the best
qualitative denoising results with finer details and less ringing artifacts.
Xiaolin Huang, Yan Xia, Lei Shi, Yixing Huang, Ming Yan, Joachim Hornegger, Andreas Maier
Subjects: Computer Vision and Pattern Recognition (cs.CV); Information Retrieval (cs.IR); Numerical Analysis (cs.NA); Numerical Analysis (math.NA)
When a measurement falls outside the quantization or measurable range, it
becomes saturated and cannot be used in classical reconstruction methods. For
example, in C-arm angiography systems, which provide projection radiography,
fluoroscopy, digital subtraction angiography, and are widely used for medical
diagnoses and interventions, the limited dynamic range of C-arm flat detectors
leads to overexposure in some projections during an acquisition, such as
imaging relatively thin body parts (e.g., the knee). Aiming at overexposure
correction for computed tomography (CT) reconstruction, we in this paper
propose a mixed one-bit compressive sensing (M1bit-CS) to acquire information
from both regular and saturated measurements. This method is inspired by the
recent progress on one-bit compressive sensing, which deals with only sign
observations. Its successful applications imply that information carried by
saturated measurements is useful to improve recovery quality. For the proposed
M1bit-CS model, alternating direction methods of multipliers is developed and
an iterative saturation detection scheme is established. Then we evaluate
M1bit-CS on one-dimensional signal recovery tasks. In some experiments, the
performance of the proposed algorithms on mixed measurements is almost the same
as recovery on unsaturated ones with the same amount of measurements. Finally,
we apply the proposed method to overexposure correction for CT reconstruction
on a phantom and a simulated clinical image. The results are promising, as the
typical streaking artifacts and capping artifacts introduced by saturated
projection data are effectively reduced, yielding significant error reduction
compared with existing algorithms based on extrapolation.
Matthias Vestner, Roee Litman, Emanuele Rodolà, Alex Bronstein, Daniel Cremers
Comments: arXiv admin note: text overlap with arXiv:1607.03425
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Many algorithms for the computation of correspondences between deformable
shapes rely on some variant of nearest neighbor matching in a descriptor space.
Such are, for example, various point-wise correspondence recovery algorithms
used as a post-processing stage in the functional correspondence framework.
Such frequently used techniques implicitly make restrictive assumptions (e.g.,
near-isometry) on the considered shapes and in practice suffer from lack of
accuracy and result in poor surjectivity. We propose an alternative recovery
technique capable of guaranteeing a bijective correspondence and producing
significantly higher accuracy and smoothness. Unlike other methods our approach
does not depend on the assumption that the analyzed shapes are isometric. We
derive the proposed method from the statistical framework of kernel density
estimation and demonstrate its performance on several challenging deformable 3D
shape matching datasets.
Xinyu Wang, Hanxi Li, Yi Li, Fumin Shen, Fatih Porikli
Comments: 6 pages
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Visual tracking is a fundamental problem in computer vision. Recently, some
deep-learning-based tracking algorithms have been achieving record-breaking
performances. However, due to the high complexity of deep learning, most deep
trackers suffer from low tracking speed, and thus are impractical in many
real-world applications. Some new deep trackers with smaller network structure
achieve high efficiency while at the cost of significant decrease on precision.
In this paper, we propose to transfer the feature for image classification to
the visual tracking domain via convolutional channel reductions. The channel
reduction could be simply viewed as an additional convolutional layer with the
specific task. It not only extracts useful information for object tracking but
also significantly increases the tracking speed. To better accommodate the
useful feature of the target in different scales, the adaptation filters are
designed with different sizes. The yielded visual tracker is real-time and also
illustrates the state-of-the-art accuracies in the experiment involving two
well-adopted benchmarks with more than 100 test videos.
Ariel Ephrat, Shmuel Peleg
Comments: Accepted for publication at ICASSP 2017
Subjects: Computer Vision and Pattern Recognition (cs.CV); Sound (cs.SD)
Speechreading is a notoriously difficult task for humans to perform. In this
paper we present an end-to-end model based on a convolutional neural network
(CNN) for generating an intelligible acoustic speech signal from silent video
frames of a speaking person. The proposed CNN generates sound features for each
frame based on its neighboring frames. Waveforms are then synthesized from the
learned speech features to produce intelligible speech. We show that by
leveraging the automatic feature learning capabilities of a CNN, we can obtain
state-of-the-art word intelligibility on the GRID dataset, and show promising
results for learning out-of-vocabulary (OOV) words.
Naoya Takahashi, Michael Gygli, Luc Van Gool
Comments: 12 pages, 9 figures. arXiv admin note: text overlap with arXiv:1604.07160
Subjects: Multimedia (cs.MM); Computer Vision and Pattern Recognition (cs.CV); Sound (cs.SD)
We propose a new deep network for audio event recognition, called AENet. In
contrast to speech, sounds coming from audio events may be produced by a wide
variety of sources. Furthermore, distinguishing them often requires analyzing
an extended time period due to the lack of clear sub-word units that are
present in speech. In order to incorporate this long-time frequency structure
of audio events, we introduce a convolutional neural network (CNN) operating on
a large temporal input. In contrast to previous works this allows us to train
an audio event detection system end-to-end. The combination of our network
architecture and a novel data augmentation outperforms previous methods for
audio event detection by 16%. Furthermore, we perform transfer learning and
show that our model learnt generic audio features, similar to the way CNNs
learn generic features on vision tasks. In video analysis, combining visual
features and traditional audio features such as MFCC typically only leads to
marginal improvements. Instead, combining visual features with our AENet
features, which can be computed efficiently on a GPU, leads to significant
performance improvements on action recognition and video highlight detection.
In video highlight detection, our audio features improve the performance by
more than 8% over visual features alone.
Wenjia Meng, Zonghua Gu, Ming Zhang, Zhaohui Wu
Subjects: Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
With the rapid proliferation of Internet of Things and intelligent edge
devices, there is an increasing need for implementing machine learning
algorithms, including deep learning, on resource-constrained mobile embedded
devices with limited memory and computation power. Typical large Convolutional
Neural Networks (CNNs) need large amounts of memory and computational power,
and cannot be deployed on embedded devices efficiently. We present Two-Bit
Networks (TBNs) for model compression of CNNs with edge weights constrained to
(-2, -1, 1, 2), which can be encoded with two bits. Our approach can reduce the
memory usage and improve computational efficiency significantly while achieving
good performance in terms of classification accuracy, thus representing a
reasonable tradeoff between model size and performance.
Karl Schlechta (LIF)
Subjects: Artificial Intelligence (cs.AI)
The elegant Stalnaker/Lewis semantics for counterfactual conditonals works
with distances between models. But human beings certainly have no tables of
models and distances in their head. We begin here an investigation using a more
realistic picture, based on findings in neuroscience. We call it a
pre-semantics, as its meaning is not a description of the world, but of the
brain, whose structure is (partly) determined by the world it reasons about.
Paul Weng
Comments: accepted at MIWAI 2017
Subjects: Artificial Intelligence (cs.AI)
In this paper, we present a link between preference-based and multiobjective
sequential decision-making. While transforming a multiobjective problem to a
preference-based one is quite natural, the other direction is a bit less
obvious. We present how this transformation (from preference-based to
multiobjective) can be done under the classic condition that preferences over
histories can be represented by additively decomposable utilities and that the
decision criterion to evaluate policies in a state is based on expectation.
This link yields a new source of multiobjective sequential decision-making
problems (i.e., when reward values are unknown) and justifies the use of
solving methods developed in one setting in the other one.
Dajian Li, Paul Weng, Orkun Karabasoglu
Comments: accepted at MIWAI 2017
Subjects: Artificial Intelligence (cs.AI)
In this paper, we tackle the problem of risk-averse route planning in a
transportation network with time-dependent and stochastic costs. To solve this
problem, we propose an adaptation of the A* algorithm that accommodates any
risk measure or decision criterion that is monotonic with first-order
stochastic dominance. We also present a case study of our algorithm on the
Manhattan, NYC, transportation network.
S. Hossein Hosseini, Tohid Nouri, Afshin Ebrahimi, S. Ali Hosseini
Comments: 6 pages, 15 figures, 1 table, IEEE International Conference on Signal Processing and Intelligent System (ICSPIS16), Dec. 2016
Subjects: Optimization and Control (math.OC); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
We propose a swarm-based optimization algorithm inspired by air currents of a
tornado. Two main air currents – spiral and updraft – are mimicked. Spiral
motion is designed for exploration of new search areas and updraft movements is
deployed for exploitation of a promising candidate solution. Assignment of just
one search direction to each particle at each iteration, leads to low
computational complexity of the proposed algorithm respect to the conventional
algorithms. Regardless of the step size parameters, the only parameter of the
proposed algorithm, called tornado diameter, can be efficiently adjusted by
randomization. Numerical results over six different benchmark cost functions
indicate comparable and, in some cases, better performance of the proposed
algorithm respect to some other metaheuristics.
Dietmar Seipel (University of Würzburg)
Comments: In Proceedings WLP’15/’16/WFLP’16, arXiv:1701.00148
Journal-ref: EPTCS 234, 2017, pp. 1-12
Subjects: Databases (cs.DB); Artificial Intelligence (cs.AI); Programming Languages (cs.PL)
Modern knowledge base systems frequently need to combine a collection of
databases in different formats: e.g., relational databases, XML databases, rule
bases, ontologies, etc. In the deductive database system DDBASE, we can manage
these different formats of knowledge and reason about them. Even the file
systems on different computers can be part of the knowledge base. Often, it is
necessary to handle different versions of a knowledge base. E.g., we might want
to find out common parts or differences of two versions of a relational
database.
We will examine the use of abstractions of rule bases by predicate dependency
and rule predicate graphs. Also the proof trees of derived atoms can help to
compare different versions of a rule base. Moreover, it might be possible to
have derivations joining rules with other formalisms of knowledge
representation.
Ontologies have shown their benefits in many applications of intelligent
systems, and there have been many proposals for rule languages compatible with
the semantic web stack, e.g., SWRL, the semantic web rule language. Recently,
ontologies are used in hybrid systems for specifying the provenance of the
different components.
Iddan Golomb, Christos Tzamos
Comments: 16 pages (3 of which are in the appendix), 4 figures
Subjects: Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI)
We address the problem of locating facilities on the ([0,1]) interval based
on reports from strategic agents. The cost of each agent is her distance to the
closest facility, and the global objective is to minimize either the maximum
cost of an agent or the social cost.
As opposed to the extensive literature on facility location which considers
the multiplicative error, we focus on minimizing the worst-case additive error.
Minimizing the additive error incentivizes mechanisms to adapt to the size of
the instance. I.e., mechanisms can sacrifice little efficiency in small
instances (location profiles in which all agents are relatively close to one
another), in order to gain more [absolute] efficiency in large instances. We
argue that this measure is better suited for many manifestations of the
facility location problem in various domains.
We present tight bounds for mechanisms locating a single facility in both
deterministic and randomized cases. We further provide several extensions for
locating multiple facilities.
Christophe Van Gysel, Evangelos Kanoulas, Maarten de Rijke
Comments: ECIR2017. Proceedings of the 39th European Conference on Information Retrieval. 2017. The final publication will be available at Springer
Subjects: Information Retrieval (cs.IR); Computation and Language (cs.CL)
We introduce pyndri, a Python interface to the Indri search engine. Pyndri
allows to access Indri indexes from Python at two levels: (1) dictionary and
tokenized document collection, (2) evaluating queries on the index. We hope
that with the release of pyndri, we will stimulate reproducible, open and
fast-paced IR research.
Xiaolin Huang, Yan Xia, Lei Shi, Yixing Huang, Ming Yan, Joachim Hornegger, Andreas Maier
Subjects: Computer Vision and Pattern Recognition (cs.CV); Information Retrieval (cs.IR); Numerical Analysis (cs.NA); Numerical Analysis (math.NA)
When a measurement falls outside the quantization or measurable range, it
becomes saturated and cannot be used in classical reconstruction methods. For
example, in C-arm angiography systems, which provide projection radiography,
fluoroscopy, digital subtraction angiography, and are widely used for medical
diagnoses and interventions, the limited dynamic range of C-arm flat detectors
leads to overexposure in some projections during an acquisition, such as
imaging relatively thin body parts (e.g., the knee). Aiming at overexposure
correction for computed tomography (CT) reconstruction, we in this paper
propose a mixed one-bit compressive sensing (M1bit-CS) to acquire information
from both regular and saturated measurements. This method is inspired by the
recent progress on one-bit compressive sensing, which deals with only sign
observations. Its successful applications imply that information carried by
saturated measurements is useful to improve recovery quality. For the proposed
M1bit-CS model, alternating direction methods of multipliers is developed and
an iterative saturation detection scheme is established. Then we evaluate
M1bit-CS on one-dimensional signal recovery tasks. In some experiments, the
performance of the proposed algorithms on mixed measurements is almost the same
as recovery on unsaturated ones with the same amount of measurements. Finally,
we apply the proposed method to overexposure correction for CT reconstruction
on a phantom and a simulated clinical image. The results are promising, as the
typical streaking artifacts and capping artifacts introduced by saturated
projection data are effectively reduced, yielding significant error reduction
compared with existing algorithms based on extrapolation.
Saeid Hosseini, Hongzhi Yin, Xiaofang Zhou, Shazia Sadiq
Subjects: Computers and Society (cs.CY); Information Retrieval (cs.IR)
Point-Of-Interest (POI) recommendation aims to mine a user’s visiting history
and find her/his potentially preferred places. Although location recommendation
methods have been studied and improved pervasively, the challenges w.r.t
employing various influences including temporal aspect still remain. Inspired
by the fact that time includes numerous granular slots (e.g. minute, hour, day,
week and etc.), in this paper, we define a new problem to perform
recommendation through exploiting all diversified temporal factors. In
particular, we argue that most existing methods only focus on a limited number
of time-related features and neglect others. Furthermore, considering a
specific granularity (e.g. time of a day) in recommendation cannot always apply
to each user or each dataset. To address the challenges, we propose a
probabilistic generative model, named after Multi-aspect Time-related Influence
(MATI) to promote POI recommendation. We also develop a novel optimization
algorithm based on Expectation Maximization (EM). Our MATI model firstly
detects a user’s temporal multivariate orientation using her check-in log in
Location-based Social Networks(LBSNs). It then performs recommendation using
temporal correlations between the user and proposed locations. Our method is
adaptable to various types of recommendation systems and can work efficiently
in multiple time-scales. Extensive experimental results on two large-scale LBSN
datasets verify the effectiveness of our method over other competitors.
Pashutan Modaresi, Philipp Gross, Siavash Sefidrodi, Mirja Eckhof, Stefan Conrad
Subjects: Computation and Language (cs.CL)
In this work, we present the results of a systematic study to investigate the
(commercial) benefits of automatic text summarization systems in a real world
scenario. More specifically, we define a use case in the context of media
monitoring and media response analysis and claim that even using a simple
query-based extractive approach can dramatically save the processing time of
the employees without significantly reducing the quality of their work.
Huijia Wu, Jiajun Zhang, Chengqing Zong
Comments: 10 pages. arXiv admin note: text overlap with arXiv:1610.03167
Subjects: Computation and Language (cs.CL)
Deep stacked RNNs are usually hard to train. Adding shortcut connections
across different layers is a common way to ease the training of stacked
networks. However, extra shortcuts make the recurrent step more complicated. To
simply the stacked architecture, we propose a framework called shortcut block,
which is a marriage of the gating mechanism and shortcuts, while discarding the
self-connected part in LSTM cell. We present extensive empirical experiments
showing that this design makes training easy and improves generalization. We
propose various shortcut block topologies and compositions to explore its
effectiveness. Based on this architecture, we obtain a 6% relatively
improvement over the state-of-the-art on CCGbank supertagging dataset. We also
get comparable results on POS tagging task.
Shi-Xiong Zhang, Zhuo Chen, Yong Zhao, Jinyu Li, Yifan Gong
Comments: @article{zhang2016End2End, title={End-to-End Attention based Text-Dependent Speaker Verification}, author={Shi-Xiong Zhang, Zhuo Chen(^{dag}), Yong Zhao, Jinyu Li and Yifan Gong}, journal={IEEE Workshop on Spoken Language Technology}, pages={171–178}, year={2016}, publisher={IEEE} }
Subjects: Computation and Language (cs.CL); Machine Learning (stat.ML)
A new type of End-to-End system for text-dependent speaker verification is
presented in this paper. Previously, using the phonetically
discriminative/speaker discriminative DNNs as feature extractors for speaker
verification has shown promising results. The extracted frame-level (DNN
bottleneck, posterior or d-vector) features are equally weighted and aggregated
to compute an utterance-level speaker representation (d-vector or i-vector). In
this work we use speaker discriminative CNNs to extract the noise-robust
frame-level features. These features are smartly combined to form an
utterance-level speaker vector through an attention mechanism. The proposed
attention model takes the speaker discriminative information and the phonetic
information to learn the weights. The whole system, including the CNN and
attention model, is joint optimized using an end-to-end criterion. The training
algorithm imitates exactly the evaluation process — directly mapping a test
utterance and a few target speaker utterances into a single verification score.
The algorithm can automatically select the most similar impostor for each
target speaker to train the network. We demonstrated the effectiveness of the
proposed end-to-end system on Windows (10) “Hey Cortana” speaker verification
task.
Peter Krejzl, Barbora Hourová, Josef Steinberger
Subjects: Computation and Language (cs.CL)
This paper describes our system created to detect stance in online
discussions. The goal is to identify whether the author of a comment is in
favor of the given target or against. Our approach is based on a maximum
entropy classifier, which uses surface-level, sentiment and domain-specific
features. The system was originally developed to detect stance in English
tweets. We adapted it to process Czech news commentaries.
Christophe Van Gysel, Evangelos Kanoulas, Maarten de Rijke
Comments: ECIR2017. Proceedings of the 39th European Conference on Information Retrieval. 2017. The final publication will be available at Springer
Subjects: Information Retrieval (cs.IR); Computation and Language (cs.CL)
We introduce pyndri, a Python interface to the Indri search engine. Pyndri
allows to access Indri indexes from Python at two levels: (1) dictionary and
tokenized document collection, (2) evaluating queries on the index. We hope
that with the release of pyndri, we will stimulate reproducible, open and
fast-paced IR research.
Dan Marsden (University of Oxford)
Comments: In Proceedings QPL 2016, arXiv:1701.00242
Journal-ref: EPTCS 236, 2017, pp. 95-107
Subjects: Logic in Computer Science (cs.LO); Computation and Language (cs.CL); Category Theory (math.CT)
We investigate notions of ambiguity and partial information in categorical
distributional models of natural language. Probabilistic ambiguity has
previously been studied using Selinger’s CPM construction. This construction
works well for models built upon vector spaces, as has been shown in quantum
computational applications. Unfortunately, it doesn’t seem to provide a
satisfactory method for introducing mixing in other compact closed categories
such as the category of sets and binary relations. We therefore lack a uniform
strategy for extending a category to model imprecise linguistic information.
In this work we adopt a different approach. We analyze different forms of
ambiguous and incomplete information, both with and without quantitative
probabilistic data. Each scheme then corresponds to a suitable enrichment of
the category in which we model language. We view different monads as
encapsulating the informational behaviour of interest, by analogy with their
use in modelling side effects in computation. Previous results of Jacobs then
allow us to systematically construct suitable bases for enrichment.
We show that we can freely enrich arbitrary dagger compact closed categories
in order to capture all the phenomena of interest, whilst retaining the
important dagger compact closed structure. This allows us to construct a model
with real convex combination of binary relations that makes non-trivial use of
the scalars. Finally we relate our various different enrichments, showing that
finite subconvex algebra enrichment covers all the effects under consideration.
Sabeur Aridhi, Alberto Montresor, Yannis Velegrakis
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC)
Recently, distributed processing of large dynamic graphs has become very
popular, especially in certain domains such as social network analysis, Web
graph analysis and spatial network analysis. In this context, many
distributed/parallel graph processing systems have been proposed, such as
Pregel, GraphLab, and Trinity. These systems can be divided into two
categories: (1) vertex-centric and (2) block-centric approaches. In
vertex-centric approaches, each vertex corresponds to a process, and message
are exchanged among vertices. In block-centric approaches, the unit of
computation is a block, a connected subgraph of the graph, and message
exchanges occur among blocks. In this paper, we are considering the issues of
scale and dynamism in the case of block-centric approaches. We present bladyg,
a block-centric framework that addresses the issue of dynamism in large-scale
graphs. We present an implementation of BLADYG on top of akka framework. We
experimentally evaluate the performance of the proposed framework.
George M Slota, Sivasankaran Rajamanickam, Kamesh Madduri
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC)
The in-memory graph layout or organization has a considerable impact on the
time and energy efficiency of distributed memory graph computations. It affects
memory locality, inter-task load balance, communication time, and overall
memory utilization. Graph layout could refer to partitioning or replication of
vertex and edge arrays, selective replication of data structures that hold
meta-data, and reordering vertex and edge identifiers. In this work, we present
DGL, a fast, parallel, and memory-efficient distributed graph layout strategy
that is specifically designed for small-world networks (low-diameter graphs
with skewed vertex degree distributions). Label propagation-based partitioning
and a scalable BFS-based ordering are the main steps in the layout strategy. We
show that the DGL layout can significantly improve end-to-end performance of
five challenging graph analytics workloads: PageRank, a parallel subgraph
enumeration program, tuned implementations of breadth-first search and
single-source shortest paths, and RDF3X-MPI, a distributed SPARQL query
processing engine. Using these benchmarks, we additionally offer a
comprehensive analysis on how graph layout affects the performance of graph
analytics with variable computation and communication characteristics.
Ema Kušen, Mark Strembeck
Subjects: Cryptography and Security (cs.CR); Distributed, Parallel, and Cluster Computing (cs.DC)
In an endeavor to reach the vision of ubiquitous computing where users are
able to use pervasive services without spatial and temporal constraints, we are
witnessing a fast growing number of mobile and sensor-enhanced devices becoming
available. However, in order to take full advantage of the numerous benefits
offered by novel mobile devices and services, we must address the related
security issues. In this paper, we present results of a systematic literature
review (SLR) on security-related topics in ubiquitous computing environments.
In our study, we found 5165 scientific contributions published between 2003 and
2015. We applied a systematic procedure to identify the threats,
vulnerabilities, attacks, as well as corresponding defense mechanisms that are
discussed in those publications. While this paper mainly discusses the results
of our study, the corresponding SLR protocol which provides all details of the
SLR is also publicly available for download.
Shuai Li
Subjects: Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
Neural networks are a revolutionary but immature technique that is fast
evolving and heavily relies on data. To benefit from the newest development and
newly available data, we want the gap between research and production as small
as possibly. On the other hand, differing from traditional machine learning
models, neural network is not just yet another statistic model, but a model for
the natural processing engine — the brain. In this work, we describe a neural
network library named { exttt akid}. It provides higher level of abstraction
for entities (abstracted as blocks) in nature upon the abstraction done on
signals (abstracted as tensors) by Tensorflow, characterizing the dataism
observation that all entities in nature processes input and emit out in some
ways. It includes a full stack of software that provides abstraction to let
researchers focus on research instead of implementation, while at the same time
the developed program can also be put into production seamlessly in a
distributed environment, and be production ready. At the top application stack,
it provides out-of-box tools for neural network applications. Lower down, akid
provides a programming paradigm that lets user easily build customized models.
The distributed computing stack handles the concurrency and communication, thus
letting models be trained or deployed to a single GPU, multiple GPUs, or a
distributed environment without affecting how a model is specified in the
programming paradigm stack. Lastly, the distributed deployment stack handles
how the distributed computing is deployed, thus decoupling the research
prototype environment with the actual production environment, and is able to
dynamically allocate computing resources, so development (Devs) and operations
(Ops) could be separated. Please refer to this http URL
for documentation.
Ibrahim Ighneiwaa, Salwa Hamidatoua, Fadia Ben Ismaela
Subjects: Learning (cs.LG); Chaotic Dynamics (nlin.CD)
Controlling Chaos could be a big factor in getting great stable amounts of
energy out of small amounts of not necessarily stable resources. By definition,
Chaos is getting huge changes in the system’s output due to unpredictable small
changes in initial conditions, and that means we could take advantage of this
fact and select the proper control system to manipulate system’s initial
conditions and inputs in general and get a desirable output out of otherwise a
Chaotic system. That was accomplished by first building some known chaotic
circuit (Chua circuit) and the NI’s MultiSim was used to simulate the ANN
control system. It was shown that this technique can also be used to stabilize
some hard to stabilize electronic systems.
Ankita Mangal, Nishant Kumar
Comments: IEEE Big Data 2016 Conference
Subjects: Learning (cs.LG)
This paper describes our approach to the Bosch production line performance
challenge run by Kaggle.com. Maximizing the production yield is at the heart of
the manufacturing industry. At the Bosch assembly line, data is recorded for
products as they progress through each stage. Data science methods are applied
to this huge data repository consisting records of tests and measurements made
for each component along the assembly line to predict internal failures. We
found that it is possible to train a model that predicts which parts are most
likely to fail. Thus a smarter failure detection system can be built and the
parts tagged likely to fail can be salvaged to decrease operating costs and
increase the profit margins.
Shuai Li
Subjects: Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
Neural networks are a revolutionary but immature technique that is fast
evolving and heavily relies on data. To benefit from the newest development and
newly available data, we want the gap between research and production as small
as possibly. On the other hand, differing from traditional machine learning
models, neural network is not just yet another statistic model, but a model for
the natural processing engine — the brain. In this work, we describe a neural
network library named { exttt akid}. It provides higher level of abstraction
for entities (abstracted as blocks) in nature upon the abstraction done on
signals (abstracted as tensors) by Tensorflow, characterizing the dataism
observation that all entities in nature processes input and emit out in some
ways. It includes a full stack of software that provides abstraction to let
researchers focus on research instead of implementation, while at the same time
the developed program can also be put into production seamlessly in a
distributed environment, and be production ready. At the top application stack,
it provides out-of-box tools for neural network applications. Lower down, akid
provides a programming paradigm that lets user easily build customized models.
The distributed computing stack handles the concurrency and communication, thus
letting models be trained or deployed to a single GPU, multiple GPUs, or a
distributed environment without affecting how a model is specified in the
programming paradigm stack. Lastly, the distributed deployment stack handles
how the distributed computing is deployed, thus decoupling the research
prototype environment with the actual production environment, and is able to
dynamically allocate computing resources, so development (Devs) and operations
(Ops) could be separated. Please refer to this http URL
for documentation.
Karamjit Singh, Garima Gupta, Lovekesh Vig, Gautam Shroff, Puneet Agarwal
Comments: Published at NIPS 2016 Workshop “What If” and Won the best Poster Award
Subjects: Learning (cs.LG)
Discovering causal models from observational and interventional data is an
important first step preceding what-if analysis or counterfactual reasoning. As
has been shown before, the direction of pairwise causal relations can, under
certain conditions, be inferred from observational data via standard
gradient-boosted classifiers (GBC) using carefully engineered statistical
features. In this paper we apply deep convolutional neural networks (CNNs) to
this problem by plotting attribute pairs as 2-D scatter plots that are fed to
the CNN as images. We evaluate our approach on the ‘Cause- Effect Pairs’ NIPS
2013 Data Challenge. We observe that a weighted ensemble of CNN with the
earlier GBC approach yields significant improvement. Further, we observe that
when less training data is available, our approach performs better than the GBC
based approach suggesting that CNN models pre-trained to determine the
direction of pairwise causal direction could have wider applicability in causal
discovery and enabling what-if or counterfactual analysis.
Wenjia Meng, Zonghua Gu, Ming Zhang, Zhaohui Wu
Subjects: Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
With the rapid proliferation of Internet of Things and intelligent edge
devices, there is an increasing need for implementing machine learning
algorithms, including deep learning, on resource-constrained mobile embedded
devices with limited memory and computation power. Typical large Convolutional
Neural Networks (CNNs) need large amounts of memory and computational power,
and cannot be deployed on embedded devices efficiently. We present Two-Bit
Networks (TBNs) for model compression of CNNs with edge weights constrained to
(-2, -1, 1, 2), which can be encoded with two bits. Our approach can reduce the
memory usage and improve computational efficiency significantly while achieving
good performance in terms of classification accuracy, thus representing a
reasonable tradeoff between model size and performance.
Morteza Ashraphijuo, Xiaodong Wang, Vaneet Aggarwal
Subjects: Information Theory (cs.IT); Learning (cs.LG); Algebraic Geometry (math.AG)
We consider the multi-view data completion problem, i.e., to complete a
matrix (mathbf{U}=[mathbf{U}_1|mathbf{U}_2]) where the ranks of
(mathbf{U},mathbf{U}_1), and (mathbf{U}_2) are given. In particular, we
investigate the fundamental conditions on the sampling pattern, i.e., locations
of the sampled entries for finite completability of such a multi-view data
given the corresponding rank constraints. In contrast with the existing
analysis on Grassmannian manifold for a single-view matrix, i.e., conventional
matrix completion, we propose a geometric analysis on the manifold structure
for multi-view data to incorporate more than one rank constraint. We provide a
deterministic necessary and sufficient condition on the sampling pattern for
finite completability. We also give a probabilistic condition in terms of the
number of samples per column that guarantees finite completability with high
probability. Finally, using the developed tools, we derive the deterministic
and probabilistic guarantees for unique completability.
Mohammad Amin Fakharian, Ashkan Esmaeili, Farokh Marvasti
Subjects: Machine Learning (stat.ML); Learning (cs.LG)
In this paper, prediction for linear systems with missing information is
investigated. New methods are introduced to improve the Mean Squared Error
(MSE) on the test set in comparison to state-of-the-art methods, through
appropriate tuning of Bias-Variance trade-off. First, the use of proposed Soft
Weighted Prediction (SWP) algorithm and its efficacy are depicted and compared
to previous works for non-missing scenarios. The algorithm is then modified and
optimized for missing scenarios. It is shown that controlled over-fitting by
suggested algorithms will improve prediction accuracy in various cases.
Simulation results approve our heuristics in enhancing the prediction accuracy.
Yeeleng Scott Vang, Xiaohui Xie
Comments: 9 pages, 3 figures, 2 tables
Subjects: Computational Engineering, Finance, and Science (cs.CE); Learning (cs.LG)
Many biological processes are governed by protein-ligand interactions. Of
such is the recognition of self and nonself cells by the immune system. This
immune response process is regulated by the major histocompatibility complex
(MHC) protein which is encoded by the human leukocyte antigen (HLA) complex.
Understanding the binding potential between MHC and peptides is crucial to our
understanding of the functioning of the immune system, which in turns will
broaden our understanding of autoimmune diseases and vaccine design.
We introduce a new distributed representation of amino acids, named HLA-Vec,
that can be used for a variety of downstream proteomic machine learning tasks.
We then propose a deep convolutional neurerror can be used only in preambleal
network architecture, named HLA-CNN, for the task of HLA class I-peptide
binding prediction. Experimental results show combining the new distributed
representation with our HLA-CNN architecture acheives state-of-the-art results
in the vast majority of the latest two Immune Epitope Database (IEDB) weekly
automated benchmark datasets.
Gonzalo H Otazu
Comments: 13 pages, 6 figures
Subjects: Numerical Analysis (cs.NA); Learning (cs.LG); Machine Learning (stat.ML)
We analyzed the performance of a biologically inspired algorithm called the
Corrected Projections Algorithm (CPA) when a sparseness constraint is required
to unambiguously reconstruct an observed signal using atoms from an
overcomplete dictionary. By changing the geometry of the estimation problem,
CPA gives an analytical expression for a binary variable that indicates the
presence or absence of a dictionary atom using an L2 regularizer. The
regularized solution can be implemented using an efficient real-time
Kalman-filter type of algorithm. The smoother L2 regularization of CPA makes it
very robust to noise, and CPA outperforms other methods in identifying known
atoms in the presence of strong novel atoms in the signal.
Masoud Arash, Ehsan Yazdian, Mohammadsadegh Fazel
Comments: 18 pages, 8 figures
Subjects: Information Theory (cs.IT)
Massive MIMO systems promise high data rates by employing large number of
antennas. By growth of number of antennas in a system, both data rate and power
usage rise. This creates an optimisation problem which specifies how many
antennas we should have for an optimum operation to achieve the best possible
Energy-Efficiency. Since the number of user terminals varies over time, number
of operational antennas should be optimised continuously while there exists a
fixed number of antennas installed in the BS. In this paper, we propose to
select appropriate number of antennas in an adaptive manner relative to number
of user terminals. Through this, existence of excessive number of antennas can
be used to select best antennas which have better channel conditions. This can
improve Energy-Efficiency due to better use of available resources. Next, we
find a tight approximation for consumed power using Wishart theorem for the
proposed scheme and use it to find a deterministic form for Energy-Efficiency,
correspond to our proposed algorithm. Our simulation results show that our
approximation is quite tight and there is significant improvement in
Energy-Efficiency when antenna selection scheme is employed.
Morteza Ashraphijuo, Xiaodong Wang, Vaneet Aggarwal
Subjects: Information Theory (cs.IT); Learning (cs.LG); Algebraic Geometry (math.AG)
We consider the multi-view data completion problem, i.e., to complete a
matrix (mathbf{U}=[mathbf{U}_1|mathbf{U}_2]) where the ranks of
(mathbf{U},mathbf{U}_1), and (mathbf{U}_2) are given. In particular, we
investigate the fundamental conditions on the sampling pattern, i.e., locations
of the sampled entries for finite completability of such a multi-view data
given the corresponding rank constraints. In contrast with the existing
analysis on Grassmannian manifold for a single-view matrix, i.e., conventional
matrix completion, we propose a geometric analysis on the manifold structure
for multi-view data to incorporate more than one rank constraint. We provide a
deterministic necessary and sufficient condition on the sampling pattern for
finite completability. We also give a probabilistic condition in terms of the
number of samples per column that guarantees finite completability with high
probability. Finally, using the developed tools, we derive the deterministic
and probabilistic guarantees for unique completability.
Chung Duc Ho, Hien Quoc Ngo, Michail Matthaiou, Trung Q. Duong
Subjects: Information Theory (cs.IT)
We consider a multi-way massive multiple-input multiple-output relay network
with zero-forcing processing at the relay. By taking into account the
time-division duplex protocol with channel estimation, we derive an analytical
approximation of the spectral efficiency. This approximation is very tight and
simple which enables us to analyze the system performance, as well as, to
compare the spectral efficiency with zero-forcing and maximum-ratio processing.
Our results show that by using a very large number of relay antennas and with
the zero-forcing technique, we can simultaneously serve many active users in
the same time-frequency resource, each with high spectral efficiency.