IT博客汇
  • 首页
  • 精华
  • 技术
  • 设计
  • 资讯
  • 扯淡
  • 权利声明
  • 登录 注册

    arXiv Paper Daily: Mon, 26 Sep 2016

    我爱机器学习(52ml.net)发表于 2016-09-26 00:00:00
    love 0

    Neural and Evolutionary Computing

    Multi-Output Artificial Neural Network for Storm Surge Prediction in North Carolina

    Anton Bezuglov, Brian Blanton, Reinaldo Santiago
    Subjects: Neural and Evolutionary Computing (cs.NE); Atmospheric and Oceanic Physics (physics.ao-ph); Applications (stat.AP)

    During hurricane seasons, emergency managers and other decision makers need
    accurate and `on-time’ information on potential storm surge impacts. Fully
    dynamical computer models, such as the ADCIRC tide, storm surge, and wind-wave
    model take several hours to complete a forecast when configured at high spatial
    resolution. Additionally, statically meaningful ensembles of high-resolution
    models (needed for uncertainty estimation) cannot easily be computed in near
    real-time. This paper discusses an artificial neural network model for storm
    surge prediction in North Carolina. The network model provides fast, real-time
    storm surge estimates at coastal locations in North Carolina. The paper studies
    the performance of the neural network model vs. other models on synthetic and
    real hurricane data.

    Deep Learning in Multi-Layer Architectures of Dense Nuclei

    Yonghua Yin, Erol Gelenbe
    Comments: 10 pages
    Subjects: Neural and Evolutionary Computing (cs.NE); Computer Vision and Pattern Recognition (cs.CV)

    In dense clusters of neurons in nuclei, cells may interconnect via
    soma-to-soma interactions, in addition to conventional synaptic connections. We
    illustrate this idea with a multi-layer architecture (MLA) composed of multiple
    clusters of recurrent sub-networks of spiking Random Neural Networks (RNN) with
    dense soma-to-soma interactions. We use this RNN-MLA architecture for deep
    learning. The inputs to the clusters are normalised by adjusting the external
    arrival rates of spikes to each cluster, and then apply this architectures to
    learning from multi-channel datasets. We present numerical results based on
    both images and sensor based data that show the value of this RNN-MLA for deep
    learning.

    Regulating Reward Training by Means of Certainty Prediction in a Neural Network-Implemented Pong Game

    Matt Oberdorfer, Matt Abuzalaf
    Comments: 7 pages, 3 figures
    Subjects: Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)

    We present the first reinforcement-learning model to self-improve its
    reward-modulated training implemented through a continuously improving
    “intuition” neural network. An agent was trained how to play the arcade video
    game Pong with two reward-based alternatives, one where the paddle was placed
    randomly during training, and a second where the paddle was simultaneously
    trained on three additional neural networks such that it could develop a sense
    of “certainty” as to how probable its own predicted paddle position will be to
    return the ball. If the agent was less than 95% certain to return the ball, the
    policy used an intuition neural network to place the paddle. We trained both
    architectures for an equivalent number of epochs and tested learning
    performance by letting the trained programs play against a near-perfect
    opponent. Through this, we found that the reinforcement learning model that
    uses an intuition neural network for placing the paddle during reward training
    quickly overtakes the simple architecture in its ability to outplay the
    near-perfect opponent, additionally outscoring that opponent by an increasingly
    wide margin after additional epochs of training.

    A Novel Progressive Multi-label Classifier for Classincremental Data

    Mihika Dave, Sahil Tapiawala, Meng Joo Er, Rajasekar Venkatesan
    Comments: 5 pages, 3 figures, 4 tables
    Subjects: Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)

    In this paper, a progressive learning algorithm for multi-label
    classification to learn new labels while retaining the knowledge of previous
    labels is designed. New output neurons corresponding to new labels are added
    and the neural network connections and parameters are automatically
    restructured as if the label has been introduced from the beginning. This work
    is the first of the kind in multi-label classifier for class-incremental
    learning. It is useful for real-world applications such as robotics where
    streaming data are available and the number of labels is often unknown. Based
    on the Extreme Learning Machine framework, a novel universal classifier with
    plug and play capabilities for progressive multi-label classification is
    developed. Experimental results on various benchmark synthetic and real
    datasets validate the efficiency and effectiveness of our proposed algorithm.


    Computer Vision and Pattern Recognition

    Real-time Human Pose Estimation from Video with Convolutional Neural Networks

    Marko Linna, Juho Kannala, Esa Rahtu
    Comments: 16 pages
    Subjects: Computer Vision and Pattern Recognition (cs.CV)

    In this paper, we present a method for real-time multi-person human pose
    estimation from video by utilizing convolutional neural networks. Our method is
    aimed for use case specific applications, where good accuracy is essential and
    variation of the background and poses is limited. This enables us to use a
    generic network architecture, which is both accurate and fast. We divide the
    problem into two phases: (1) pre-training and (2) finetuning. In pre-training,
    the network is learned with highly diverse input data from publicly available
    datasets, while in finetuning we train with application specific data, which we
    record with Kinect. Our method differs from most of the state-of-the-art
    methods in that we consider the whole system, including person detector, pose
    estimator and an automatic way to record application specific training material
    for finetuning. Our method is considerably faster than many of the
    state-of-the-art methods. Our method can be thought of as a replacement for
    Kinect, and it can be used for higher level tasks, such as gesture control,
    games, person tracking, action recognition and action tracking. We achieved
    accuracy of 96.8\% (PCK@0.2) with application specific data.

    The face-space duality hypothesis: a computational model

    Jonathan Vitale, Mary-Anne Williams, Benjamin Johnston
    Comments: in 38th Annual Meeting of the Cognitive Science Society, 2016
    Subjects: Computer Vision and Pattern Recognition (cs.CV)

    Valentine’s face-space suggests that faces are represented in a psychological
    multidimensional space according to their perceived properties. However, the
    proposed framework was initially designed as an account of invariant facial
    features only, and explanations for dynamic features representation were
    neglected. In this paper we propose, develop and evaluate a computational model
    for a twofold structure of the face-space, able to unify both identity and
    expression representations in a single implemented model. To capture both
    invariant and dynamic facial features we introduce the face-space duality
    hypothesis and subsequently validate it through a mathematical presentation
    using a general approach to dimensionality reduction. Two experiments with real
    facial images show that the proposed face-space: (1) supports both identity and
    expression recognition, and (2) has a twofold structure anticipated by our
    formal argument.

    Example-Based Image Synthesis via Randomized Patch-Matching

    Yi Ren, Yaniv Romano, Michael Elad
    Subjects: Computer Vision and Pattern Recognition (cs.CV)

    Image and texture synthesis is a challenging task that has long been drawing
    attention in the fields of image processing, graphics, and machine learning.
    This problem consists of modelling the desired type of images, either through
    training examples or via a parametric modeling, and then generating images that
    belong to the same statistical origin.

    This work addresses the image synthesis task, focusing on two specific
    families of images — handwritten digits and face images. This paper offers two
    main contributions. First, we suggest a simple and intuitive algorithm capable
    of generating such images in a unified way. The proposed approach taken is
    pyramidal, consisting of upscaling and refining the estimated image several
    times. For each upscaling stage, the algorithm randomly draws small patches
    from a patch database, and merges these to form a coherent and novel image with
    high visual quality. The second contribution is a general framework for the
    evaluation of the generation performance, which combines three aspects: the
    likelihood, the originality and the spread of the synthesized images. We assess
    the proposed synthesis scheme and show that the results are similar in nature,
    and yet different from the ones found in the training set, suggesting that true
    synthesis effect has been obtained.

    EgoCap: Egocentric Marker-less Motion Capture with Two Fisheye Cameras

    Helge Rhodin, Christian Richardt, Dan Casas, Eldar Insafutdinov, Mohammad Shafiei, Hans-Peter Seidel, Bernt Schiele, Christian Theobalt
    Comments: SIGGRAPH Asia 2016
    Subjects: Computer Vision and Pattern Recognition (cs.CV)

    Marker-based and marker-less optical skeletal motion-capture methods use an
    outside-in arrangement of cameras placed around a scene, with viewpoints
    converging on the center. They often create discomfort by possibly needed
    marker suits, and their recording volume is severely restricted and often
    constrained to indoor scenes with controlled backgrounds. Alternative
    suit-based systems use several inertial measurement units or an exoskeleton to
    capture motion. This makes capturing independent of a confined volume, but
    requires substantial, often constraining, and hard to set up body
    instrumentation. We therefore propose a new method for real-time, marker-less
    and egocentric motion capture which estimates the full-body skeleton pose from
    a lightweight stereo pair of fisheye cameras that are attached to a helmet or
    virtual reality headset. It combines the strength of a new generative pose
    estimation framework for fisheye views with a ConvNet-based body-part detector
    trained on a large new dataset. Our inside-in method captures full-body motion
    in general indoor and outdoor scenes, and also crowded scenes with many people
    in close vicinity. The captured user can freely move around, which enables
    reconstruction of larger-scale activities and is particularly useful in virtual
    reality to freely roam and interact, while seeing the fully motion-captured
    virtual body.

    Funnel-Structured Cascade for Multi-View Face Detection with Alignment-Awareness

    Shuzhe Wu, Meina Kan, Zhenliang He, Shiguang Shan, Xilin Chen
    Comments: Submitted to Neurocomputing (under review). An adapted open source implementation can be found at this https URL
    Subjects: Computer Vision and Pattern Recognition (cs.CV)

    Multi-view face detection in open environment is a challenging task due to
    diverse variations of face appearances and shapes. Most multi-view face
    detectors depend on multiple models and organize them in parallel, pyramid or
    tree structure, which compromise between the accuracy and time-cost. Aiming at
    a more favorable multi-view face detector, we propose a novel funnel-structured
    cascade (FuSt) detection framework. In a coarse-to-fine flavor, our FuSt
    consists of, from top to bottom, 1) multiple view-specific fast LAB cascade for
    extremely quick face proposal, 2) multiple coarse MLP cascade for further
    candidate window verification, and 3) a unified fine MLP cascade with
    shape-indexed features for accurate face detection. Compared with other
    structures, on the one hand, the proposed one uses multiple computationally
    efficient distributed classifiers to propose a small number of candidate
    windows but with a high recall of multi-view faces. On the other hand, by using
    a unified MLP cascade to examine proposals of all views in a centralized style,
    it provides a favorable solution for multi-view face detection with high
    accuracy and low time-cost. Besides, the FuSt detector is alignment-aware and
    performs a coarse facial part prediction which is beneficial for subsequent
    face alignment. Extensive experiments on two challenging datasets, FDDB and
    AFW, demonstrate the effectiveness of our FuSt detector in both accuracy and
    speed.

    EFANNA : An Extremely Fast Approximate Nearest Neighbor Search Algorithm Based on kNN Graph

    Cong Fu, Deng Cai
    Subjects: Computer Vision and Pattern Recognition (cs.CV)

    Approximate nearest neighbor (ANN) search is a fundamental problem in many
    areas of data mining, machine learning and computer vision. The performance of
    traditional hierarchical structure (tree) based methods decreases as the
    dimensionality of data grows, while hashing based methods usually lack
    efficiency in practice. Recently, the graph based methods have drawn
    considerable attention. The main idea is that emph{a neighbor of a neighbor is
    also likely to be a neighbor}, which we refer as emph{NN-expansion}. These
    methods construct a $k$-nearest neighbor ($k$NN) graph offline. And at online
    search stage, these methods find candidate neighbors of a query point in some
    way (eg, random selection), and then check the neighbors of these candidate
    neighbors for closer ones iteratively. Despite some promising results, there
    are mainly two problems with these approaches: 1) These approaches tend to
    converge to local optima. 2) Constructing a $k$NN graph is time consuming. We
    find that these two problems can be nicely solved when we provide a good
    initialization for NN-expansion. In this paper, we propose EFANNA, an extremely
    fast approximate nearest neighbor search algorithm based on $k$NN Graph. Efanna
    nicely combines the advantages of hierarchical structure based methods and
    nearest-neighbor-graph based methods. Extensive experiments have shown that
    EFANNA outperforms the state-of-art algorithms both on approximate nearest
    neighbor search and approximate nearest neighbor graph construction. To the
    best of our knowledge, EFANNA is the fastest algorithm so far both on
    approximate nearest neighbor graph construction and approximate nearest
    neighbor search. A library EFANNA based on this research is released on Github.

    Deep Quality: A Deep No-reference Quality Assessment System

    Prajna Paramita Dash, Akshaya Mishra, Alexander Wong
    Comments: 2 pages
    Subjects: Multimedia (cs.MM); Computer Vision and Pattern Recognition (cs.CV)

    Image quality assessment (IQA) continues to garner great interest in the
    research community, particularly given the tremendous rise in consumer video
    capture and streaming. Despite significant research effort in IQA in the past
    few decades, the area of no-reference image quality assessment remains a great
    challenge and is largely unsolved. In this paper, we propose a novel
    no-reference image quality assessment system called Deep Quality, which
    leverages the power of deep learning to model the complex relationship between
    visual content and the perceived quality. Deep Quality consists of a novel
    multi-scale deep convolutional neural network, trained to learn to assess image
    quality based on training samples consisting of different distortions and
    degradations such as blur, Gaussian noise, and compression artifacts.
    Preliminary results using the CSIQ benchmark image quality dataset showed that
    Deep Quality was able to achieve strong quality prediction performance (89%
    patch-level and 98% image-level prediction accuracy), being able to achieve
    similar performance as full-reference IQA methods.

    Deep Learning in Multi-Layer Architectures of Dense Nuclei

    Yonghua Yin, Erol Gelenbe
    Comments: 10 pages
    Subjects: Neural and Evolutionary Computing (cs.NE); Computer Vision and Pattern Recognition (cs.CV)

    In dense clusters of neurons in nuclei, cells may interconnect via
    soma-to-soma interactions, in addition to conventional synaptic connections. We
    illustrate this idea with a multi-layer architecture (MLA) composed of multiple
    clusters of recurrent sub-networks of spiking Random Neural Networks (RNN) with
    dense soma-to-soma interactions. We use this RNN-MLA architecture for deep
    learning. The inputs to the clusters are normalised by adjusting the external
    arrival rates of spikes to each cluster, and then apply this architectures to
    learning from multi-channel datasets. We present numerical results based on
    both images and sensor based data that show the value of this RNN-MLA for deep
    learning.


    Artificial Intelligence

    Regulating Reward Training by Means of Certainty Prediction in a Neural Network-Implemented Pong Game

    Matt Oberdorfer, Matt Abuzalaf
    Comments: 7 pages, 3 figures
    Subjects: Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)

    We present the first reinforcement-learning model to self-improve its
    reward-modulated training implemented through a continuously improving
    “intuition” neural network. An agent was trained how to play the arcade video
    game Pong with two reward-based alternatives, one where the paddle was placed
    randomly during training, and a second where the paddle was simultaneously
    trained on three additional neural networks such that it could develop a sense
    of “certainty” as to how probable its own predicted paddle position will be to
    return the ball. If the agent was less than 95% certain to return the ball, the
    policy used an intuition neural network to place the paddle. We trained both
    architectures for an equivalent number of epochs and tested learning
    performance by letting the trained programs play against a near-perfect
    opponent. Through this, we found that the reinforcement learning model that
    uses an intuition neural network for placing the paddle during reward training
    quickly overtakes the simple architecture in its ability to outplay the
    near-perfect opponent, additionally outscoring that opponent by an increasingly
    wide margin after additional epochs of training.

    Optimizing positional scoring rules for rank aggregation

    Ioannis Caragiannis, Xenophon Chatzigeorgiou, George A. Krimpas, Alexandros A. Voudouris
    Subjects: Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI); Computational Complexity (cs.CC)

    Nowadays, several crowdsourcing projects exploit social choice methods for
    computing an aggregate ranking of alternatives given individual rankings
    provided by workers. Motivated by such systems, we consider a setting where
    each worker is asked to rank a fixed (small) number of alternatives and, then,
    a positional scoring rule is used to compute the aggregate ranking. Among the
    apparently infinite such rules, what is the best one to use? To answer this
    question, we assume that we have partial access to an underlying true ranking.
    Then, the important optimization problem to be solved is to compute the
    positional scoring rule whose outcome, when applied to the profile of
    individual rankings, is as close as possible to the part of the underlying true
    ranking we know. We study this fundamental problem from a theoretical viewpoint
    and present positive and negative complexity results and, furthermore,
    complement our theoretical findings with experiments on real-world and
    synthetic data.

    Discovering Sound Concepts and Acoustic Relations In Text

    Anurag Kumar, Bhiksha Raj, Ndapandula Nakashole
    Comments: 5 pages
    Subjects: Sound (cs.SD); Artificial Intelligence (cs.AI); Learning (cs.LG)

    In this paper we describe approaches for discovering acoustic concepts and
    relations in text. The first major goal is to be able to identify text phrases
    which contain a notion of audibility and can be termed as a sound or an
    acoustic concept. We also propose a method to define an acoustic scene through
    a set of sound concepts. We use pattern matching and parts of speech tags to
    generate sound concepts from large scale text corpora. We use dependency
    parsing and LSTM recurrent neural network to predict a set of sound concepts
    for a given acoustic scene. These methods are not only helpful in creating an
    acoustic knowledge base but also directly help in acoustic event and scene
    detection research in a variety of ways.

    Towards the bio-personalization of music recommendation systems: A single-sensor EEG biomarker of subjective music preference

    Dimitrios A. Adamos (1 and 3), Stavros I. Dimitriadis (2), Nikolaos A. Laskaris (2 and 3), ((1) School of Music Studies, Faculty of Fine Arts, Aristotle University of Thessaloniki, (2) AIIA Lab, Department of Informatics, Aristotle University of Thessaloniki, (3) Neuroinformatics GRoup, Aristotle University of Thessaloniki)
    Journal-ref: Information Sciences, Volumes 343 – 344, 20 May 2016, Pages 94 –
    108
    Subjects: Neurons and Cognition (q-bio.NC); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Multimedia (cs.MM)

    Recent advances in biosensors technology and mobile electroencephalographic
    (EEG) interfaces have opened new application fields for cognitive monitoring. A
    computable biomarker for the assessment of spontaneous aesthetic brain
    responses during music listening is introduced here. It derives from
    well-established measures of cross-frequency coupling (CFC) and quantifies the
    music-induced alterations in the dynamic relationships between brain rhythms.
    During a stage of exploratory analysis, and using the signals from a suitably
    designed experiment, we established the biomarker, which acts on brain
    activations recorded over the left prefrontal cortex and focuses on the
    functional coupling between high-beta and low-gamma oscillations. Based on data
    from an additional experimental paradigm, we validated the introduced biomarker
    and showed its relevance for expressing the subjective aesthetic appreciation
    of a piece of music. Our approach resulted in an affordable tool that can
    promote human-machine interaction and, by serving as a personalized music
    annotation strategy, can be potentially integrated into modern flexible music
    recommendation systems.

    Keywords: Cross-frequency coupling; Human-computer interaction;
    Brain-computer interface

    Language as a Latent Variable: Discrete Generative Models for Sentence Compression

    Yishu Miao, Phil Blunsom
    Comments: EMNLP 2016
    Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)

    In this work we explore deep generative models of text in which the latent
    representation of a document is itself drawn from a discrete language model
    distribution. We formulate a variational auto-encoder for inference in this
    model and apply it to the task of compressing sentences. In this application
    the generative model first draws a latent summary sentence from a background
    language model, and then subsequently draws the observed sentence conditioned
    on this latent summary. In our empirical evaluation we show that generative
    formulations of both abstractive and extractive compression yield
    state-of-the-art results when trained on a large amount of supervised data.
    Further, we explore semi-supervised compression scenarios where we show that it
    is possible to achieve performance competitive with previously proposed
    supervised models while training on a fraction of the supervised data.


    Computation and Language

    Incorporating Relation Paths in Neural Relation Extraction

    Wenyuan Zeng, Yankai Lin, Zhiyuan Liu, Maosong Sun
    Comments: 9 pages, 3 figures, 4 tables
    Subjects: Computation and Language (cs.CL)

    Distantly supervised relation extraction has been widely used to find novel
    relational facts from plain text. To predict the relation between a pair of two
    target entities, existing methods solely rely on those direct sentences
    containing both entities. In fact, there are also many sentences containing
    only one of the target entities, which provide rich and useful information for
    relation extraction. To address this issue, we build inference chains between
    two target entities via intermediate entities, and propose a path-based neural
    relation extraction model to encode the relational semantics from both direct
    sentences and inference chains. Experimental results on real-world datasets
    show that, our model can make full use of those sentences containing only one
    target entity, and achieves significant and consistent improvements on relation
    extraction as compared with baselines.

    AMR-to-text generation as a Traveling Salesman Problem

    Linfeng Song, Yue Zhang, Xiaochang Peng, Zhiguo Wang, Daniel Gildea
    Comments: accepted by EMNLP 2016
    Subjects: Computation and Language (cs.CL)

    The task of AMR-to-text generation is to generate grammatical text that
    sustains the semantic meaning for a given AMR graph. We at- tack the task by
    first partitioning the AMR graph into smaller fragments, and then generating
    the translation for each fragment, before finally deciding the order by solving
    an asymmetric generalized traveling salesman problem (AGTSP). A Maximum Entropy
    classifier is trained to estimate the traveling costs, and a TSP solver is used
    to find the optimized solution. The final model reports a BLEU score of 22.44
    on the SemEval-2016 Task8 dataset.

    Language as a Latent Variable: Discrete Generative Models for Sentence Compression

    Yishu Miao, Phil Blunsom
    Comments: EMNLP 2016
    Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)

    In this work we explore deep generative models of text in which the latent
    representation of a document is itself drawn from a discrete language model
    distribution. We formulate a variational auto-encoder for inference in this
    model and apply it to the task of compressing sentences. In this application
    the generative model first draws a latent summary sentence from a background
    language model, and then subsequently draws the observed sentence conditioned
    on this latent summary. In our empirical evaluation we show that generative
    formulations of both abstractive and extractive compression yield
    state-of-the-art results when trained on a large amount of supervised data.
    Further, we explore semi-supervised compression scenarios where we show that it
    is possible to achieve performance competitive with previously proposed
    supervised models while training on a fraction of the supervised data.

    Deep Multi-Task Learning with Shared Memory

    Pengfei Liu, Xipeng Qiu, Xuanjing Huang
    Comments: accepted by emnlp2016. arXiv admin note: text overlap with arXiv:1605.05101
    Subjects: Computation and Language (cs.CL)

    Neural network based models have achieved impressive results on various
    specific tasks. However, in previous works, most models are learned separately
    based on single-task supervised objectives, which often suffer from
    insufficient training data. In this paper, we propose two deep architectures
    which can be trained jointly on multiple related tasks. More specifically, we
    augment neural model with an external memory, which is shared by several tasks.
    Experiments on two groups of text classification tasks show that our proposed
    architectures can improve the performance of a task with the help of other
    related tasks.

    Annotating Derivations: A New Evaluation Strategy and Dataset for Algebra Word Problems

    Shyam Upadhyay, Ming-Wei Chang
    Subjects: Computation and Language (cs.CL)

    We propose a new evaluation for automatic solvers for algebra word problems,
    which can identify reasoning mistakes that existing evaluations overlook. Our
    proposal is to use derivations for evaluations, which reflect the reasoning
    process of the solver by explaining how the equation system was constructed. We
    accomplish this by developing an algorithm for checking the equivalence between
    two derivations, and showing how derivation annotations can be
    semi-automatically added to existing datasets. To make our experiments more
    comprehensive, we also annotated DRAW-1K , a new dataset of 1000 general
    algebra word problems. In total, our experiments span over 2300 algebra word
    problems. We found that the annotated derivation enable a superior evaluation
    of automatic solvers than previously used metrics.

    Novel stochastic properties of the short-time spectrum for unvoiced pronunciation modeling and synthesis

    Xiaodong Zhuang, Nikos E. Mastorakis
    Comments: 24 pages, 16 figures, original work
    Subjects: Sound (cs.SD); Computation and Language (cs.CL)

    Stochastic property of speech signal is a fundamental research topic in
    speech analysis and processing. In this paper, multiple levels of randomness in
    speech signal are discussed, and the stochastic properties of unvoiced
    pronunciation are studied in detail, which has not received sufficient research
    attention before. The study is based on the signals of sustained unvoiced
    pronunciation captured in the experiments, for which the amplitude and phase
    values in the short-time spectrum are studied as random variables. The
    statistics of amplitude for each frequency component is studied individually,
    based on which a new property of “consistent standard deviation coefficient” is
    revealed for the amplitude spectrum of unvoiced pronunciation. The relationship
    between the amplitude probability distributions of different frequency
    components is further studied, which indicates that all the frequency
    components have a common prototype of amplitude probability distribution. As an
    adaptive and flexible probability distribution, the Weibull distribution is
    adopted to fit the expectation-normalized amplitude spectrum data. The phase
    distribution for the short-time spectrum is also studied, and the results show
    a uniform distribution. A synthesis method for unvoiced pronunciation is
    proposed based on the Weibull distribution of amplitude and uniform
    distribution of phase, which is implemented by STFT with artificially generated
    short-time spectrum with random amplitude and phase. The synthesis results have
    identical quality of auditory perception as the original pronunciation, and
    have similar autocorrelation as that of the original signal, which proves the
    effectiveness of the proposed stochastic model of short-time spectrum for
    unvoiced pronunciation.


    Distributed, Parallel, and Cluster Computing

    MPI Parallelization of the Resistive Wall Code STARWALL: Report of the EUROfusion High Level Support Team Project JORSTAR

    S. Mochalskyy, M. Hoelzl, R. Hatzky
    Subjects: Distributed, Parallel, and Cluster Computing (cs.DC)

    Large scale plasma instabilities inside a tokamak can be influenced by the
    currents flowing in the conducting vessel wall. This involves non linear plasma
    dynamics and its interaction with the wall current. In order to study this
    problem the code that solves the magneto-hydrodynamic (MHD) equations, called
    JOREK, was coupled with the model for the vacuum region and the resistive
    conducting structure named STARWALL. The JOREK-STARWALL model has been already
    applied to perform simulations of the Vertical Displacement Events (VDEs), the
    Resistive Wall Modes (RWMs), and Quiescent H-Mode.

    At the beginning of the project it was not possible to resolve the realistic
    wall structure with a large number of finite element triangles due to the huge
    consumption of memory and wall clock time by STARWALL and the corresponding
    coupling routine in JOREK. Moreover, both the STARWALL code and the JOREK
    coupling routine are only partially parallelized via OpenMP. The aim of this
    project is to implement an MPI parallelization in the model that should allow
    to obtain realistic results with high resolution. This project concentrates on
    the MPI parallelization of STARWALL. Parallel I/O and the MPI parallelization
    of the coupling terms inside JOREK will be addressed in a follow-up project.

    Scheduling Under Power and Energy Constraints

    Mohammed Haroon Dupty, Pragati Agrawal, Shrisha Rao
    Comments: 28 pages
    Subjects: Data Structures and Algorithms (cs.DS); Distributed, Parallel, and Cluster Computing (cs.DC)

    Given a system model where machines have distinct speeds and power ratings
    but are otherwise compatible, we consider various problems of scheduling under
    resource constraints on the system which place the restriction that not all
    machines can be run at once. These can be power, energy, or makespan
    constraints on the system. Given such constraints, there are problems with
    divisible as well as non-divisible jobs. In the setting where there is a
    constraint on power, we show that the problem of minimizing makespan for a set
    of divisible jobs is NP-hard by reduction to the knapsack problem. We then show
    that scheduling to minimize energy with power constraints is also NP-hard. We
    then consider scheduling with energy and makespan constraints with divisible
    jobs and show that these can be solved in polynomial time, and the problems
    with non-divisible jobs are NP-hard. We give exact and approximation algorithms
    for these problems as required.

    Hydra: Leveraging Functional Slicing for Efficient Distributed SDN Controllers

    Yiyang Chang, Ashkan Rezaei, Balajee Vamanan, Jahangir Hasan, Sanjay Rao, T. N. Vijaykumar
    Comments: 8 pages
    Subjects: Networking and Internet Architecture (cs.NI); Distributed, Parallel, and Cluster Computing (cs.DC)

    The conventional approach to scaling Software Defined Networking (SDN)
    controllers today is to partition switches based on network topology, with each
    partition being controlled by a single physical controller, running all SDN
    applications. However, topological partitioning is limited by the fact that (i)
    performance of latency-sensitive (e.g., monitoring) SDN applications associated
    with a given partition may be impacted by co-located compute-intensive (e.g.,
    route computation) applications; (ii) simultaneously achieving low convergence
    time and response times might be challenging; and (iii) communication between
    instances of an application across partitions may increase latencies. To tackle
    these issues, in this paper, we explore functional slicing, a complementary
    approach to scaling, where multiple SDN applications belonging to the same
    topological partition may be placed in physically distinct servers. We present
    Hydra, a framework for distributed SDN controllers based on functional slicing.
    Hydra chooses partitions based on convergence time as the primary metric, but
    places application instances across partitions in a manner that keeps response
    times low while considering communication between applications of a partition,
    and instances of an application across partitions. Evaluations using the
    Floodlight controller show the importance and effectiveness of Hydra in
    simultaneously keeping convergence times on failures small, while sustaining
    higher throughput per partition and ensuring responsiveness to
    latency-sensitive applications.


    Learning

    Using Neural Network Formalism to Solve Multiple-Instance Problems

    Tomas Pevny, Petr Somol
    Subjects: Learning (cs.LG); Machine Learning (stat.ML)

    Many objects in the real world are difficult to describe by a single
    numerical vector of a fixed length, whereas describing them by a set of vectors
    is more natural. Therefore, Multiple instance learning (MIL) techniques have
    been constantly gaining on importance throughout last years. MIL formalism
    represents each object (sample) by a set (bag) of feature vectors (instances)
    of fixed length where knowledge about objects (e.g., class label) is available
    on bag level but not necessarily on instance level. Many standard tools
    including supervised classifiers have been already adapted to MIL setting since
    the problem got formalized in late nineties. In this work we propose a neural
    network (NN) based formalism that intuitively bridges the gap between MIL
    problem definition and the vast existing knowledge-base of standard models and
    classifiers. We show that the proposed NN formalism is effectively optimizable
    by a modified back-propagation algorithm and can reveal unknown patterns inside
    bags. Comparison to eight types of classifiers from the prior art on a set of
    14 publicly available benchmark datasets confirms the advantages and accuracy
    of the proposed solution.

    A Novel Progressive Multi-label Classifier for Classincremental Data

    Mihika Dave, Sahil Tapiawala, Meng Joo Er, Rajasekar Venkatesan
    Comments: 5 pages, 3 figures, 4 tables
    Subjects: Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)

    In this paper, a progressive learning algorithm for multi-label
    classification to learn new labels while retaining the knowledge of previous
    labels is designed. New output neurons corresponding to new labels are added
    and the neural network connections and parameters are automatically
    restructured as if the label has been introduced from the beginning. This work
    is the first of the kind in multi-label classifier for class-incremental
    learning. It is useful for real-world applications such as robotics where
    streaming data are available and the number of labels is often unknown. Based
    on the Extreme Learning Machine framework, a novel universal classifier with
    plug and play capabilities for progressive multi-label classification is
    developed. Experimental results on various benchmark synthetic and real
    datasets validate the efficiency and effectiveness of our proposed algorithm.

    Multilayer Spectral Graph Clustering via Convex Layer Aggregation

    Pin-Yu Chen, Alfred O. Hero III
    Comments: To appear at IEEE GlobalSIP 2016
    Subjects: Learning (cs.LG); Social and Information Networks (cs.SI); Machine Learning (stat.ML)

    Multilayer graphs are commonly used for representing different relations
    between entities and handling heterogeneous data processing tasks. New
    challenges arise in multilayer graph clustering for assigning clusters to a
    common multilayer node set and for combining information from each layer. This
    paper presents a theoretical framework for multilayer spectral graph clustering
    of the nodes via convex layer aggregation. Under a novel multilayer signal plus
    noise model, we provide a phase transition analysis that establishes the
    existence of a critical value on the noise level that permits reliable cluster
    separation. The analysis also specifies analytical upper and lower bounds on
    the critical value, where the bounds become exact when the clusters have
    identical sizes. Numerical experiments on synthetic multilayer graphs are
    conducted to validate the phase transition analysis and study the effect of
    layer weights and noise levels on clustering reliability.

    Input Convex Neural Networks

    Brandon Amos, Lei Xu, J. Zico Kolter
    Subjects: Learning (cs.LG); Optimization and Control (math.OC)

    This paper presents the input convex neural network architecture. These are
    scalar-valued (potentially deep) neural networks with constraints on the
    network parameters such that the output of the network is a convex function of
    (some of) the inputs. The networks allow for efficient inference via
    optimization over some inputs to the network given others, and can be applied
    to settings including structured prediction, data imputation, reinforcement
    learning, and others. In this paper we lay the basic groundwork for these
    models, proposing methods for inference, optimization and learning, and analyze
    their representational power. We show that many existing neural network
    architectures can be made input-convex with only minor modification, and
    develop specialized optimization algorithms tailored to this setting. Finally,
    we highlight the performance of the methods on multi-label prediction, image
    completion, and reinforcement learning problems, where we show improvement over
    the existing state of the art in many cases.

    Screening Rules for Convex Problems

    Anant Raj, Jakob Olbrich, Bernd Gärtner, Bernhard Schölkopf, Martin Jaggi
    Subjects: Optimization and Control (math.OC); Learning (cs.LG); Machine Learning (stat.ML)

    We propose a new framework for deriving screening rules for convex
    optimization problems. Our approach covers a large class of constrained and
    penalized optimization formulations, and works in two steps. First, given any
    approximate point, the structure of the objective function and the duality gap
    is used to gather information on the optimal solution. In the second step, this
    information is used to produce screening rules, i.e. safely identifying
    unimportant weight variables of the optimal solution. Our general framework
    leads to a large variety of useful existing as well as new screening rules for
    many applications. For example, we provide new screening rules for general
    simplex and $L_1$-constrained problems, Elastic Net, squared-loss Support
    Vector Machines, minimum enclosing ball, as well as structured norm regularized
    problems, such as group lasso.

    Gated Neural Networks for Option Pricing: Rationality by Design

    Yongxin Yang, Yu Zheng, Timothy M. Hospedales
    Comments: Technical Report. 7 pages, 5 figures
    Subjects: Computational Finance (q-fin.CP); Learning (cs.LG); Pricing of Securities (q-fin.PR)

    We propose a neural network approach to price EU call options that
    significantly outperforms some existing pricing models and comes with
    guarantees that its predictions are economically reasonable. To achieve this,
    we introduce a class of gated neural networks that automatically learn to
    divide-and-conquer the problem space for robust and accurate pricing. We then
    derive instantiations of these networks that are ‘rational by design’ in terms
    of naturally encoding a valid call option surface that enforces no arbitrage
    principles. This integration of human insight within data-driven learning
    provides significantly better generalisation in pricing performance due to the
    encoded inductive bias in the learning, guarantees sanity in the model’s
    predictions, and provides econometrically useful byproduct such as risk neutral
    density.

    Discovering Sound Concepts and Acoustic Relations In Text

    Anurag Kumar, Bhiksha Raj, Ndapandula Nakashole
    Comments: 5 pages
    Subjects: Sound (cs.SD); Artificial Intelligence (cs.AI); Learning (cs.LG)

    In this paper we describe approaches for discovering acoustic concepts and
    relations in text. The first major goal is to be able to identify text phrases
    which contain a notion of audibility and can be termed as a sound or an
    acoustic concept. We also propose a method to define an acoustic scene through
    a set of sound concepts. We use pattern matching and parts of speech tags to
    generate sound concepts from large scale text corpora. We use dependency
    parsing and LSTM recurrent neural network to predict a set of sound concepts
    for a given acoustic scene. These methods are not only helpful in creating an
    acoustic knowledge base but also directly help in acoustic event and scene
    detection research in a variety of ways.

    Constraint-Based Clustering Selection

    Toon Van Craenendonck, Hendrik Blockeel
    Subjects: Machine Learning (stat.ML); Learning (cs.LG)

    Semi-supervised clustering methods incorporate a limited amount of
    supervision into the clustering process. Typically, this supervision is
    provided by the user in the form of pairwise constraints. Existing methods use
    such constraints in one of the following ways: they adapt their clustering
    procedure, their similarity metric, or both. All of these approaches operate
    within the scope of individual clustering algorithms. In contrast, we propose
    to use constraints to choose between clusterings generated by very different
    unsupervised clustering algorithms, run with different parameter settings. We
    empirically show that this simple approach often outperforms existing
    semi-supervised clustering methods.


    Information Theory

    Channel Estimation and Performance Analysis of One-Bit Massive MIMO Systems

    Yongzhi Li, Cheng Tao, Gonzalo Seco-Granados, Amine Mezghani, A. Lee Swindlehurst, Liu Liu
    Comments: 14 pages, 9 figures, submitted to IEEE Trans. on Signal Processing
    Subjects: Information Theory (cs.IT)

    This paper considers channel estimation and system performance for the uplink
    of a single-cell massive multiple-input multiple-output (MIMO) system. Each
    receive antenna of the base station (BS) is assumed to be equipped with a pair
    of one-bit analog-to-digital converters (ADCs) to quantize the real and
    imaginary part of the received signal. We first obtain the Cramer-Rao lower
    bound for channel estimation, and show that the one-bit ADCs cause the
    performance of unbiased channel estimators to degrade at high SNR. We then
    propose a biased channel estimator based on the Bussgang decomposition, which
    reformulates the nonlinear quantizer as a linear function with identical first-
    and second-order statistics. The resulting channel estimator works well at high
    SNRs and outperforms previously proposed approaches across all SNRs.We then
    derive closed-form expressions at low SNR for an approximation of the
    achievable rate for the maximal ratio combiner and zero forcing receivers that
    takes channel estimation error due to both noise and one bit quantization into
    account. The closed-form expressions in turn allow us to obtain insight into
    important system design issues such as optimal resource allocation, maximal sum
    spectral efficiency, overall energy efficiency, and number of antennas.
    Numerical results are presented to verify our analytical results and
    demonstrate the benefit of optimizing system performance accordingly.

    An extended characterization of a class of optimal three-weight cyclic codes over any finite field

    Gerardo Vega
    Comments: arXiv admin note: text overlap with arXiv:1508.05077
    Subjects: Information Theory (cs.IT)

    A characterization of a class of optimal three-weight cyclic codes of
    dimension 3 over any finite field was recently presented in [10]. Shortly after
    this, a generalization for the sufficient numerical conditions of such
    characterization was given in [3]. The main purpose of this work is to show
    that the numerical conditions found in [3], are also necessary. As we will see
    later, an interesting feature of the present work, in clear contrast with these
    two preceding works, is that we use some new and non-conventional methods in
    order to achieve our goals. In fact, through these non-conventional methods, we
    not only were able to extend the characterization in [10], but also present a
    less complex proof of such extended characterization, which avoids the use of
    some of the sophisticated –but at the same time complex– theorems, that are
    the key arguments of the proofs given in [10] and [3]. Furthermore, we also
    find the parameters for the dual code of any cyclic code in our extended
    characterization class. In fact, after the analysis of some examples, it seems
    that such dual codes always have the same parameters as the best known linear
    codes.

    On the Non-Existence of Unbiased Estimators in Constrained Estimation Problems

    Anelia Somekh-Baruch, Amir Leshem, Venkatesh Saligrama
    Subjects: Statistics Theory (math.ST); Information Theory (cs.IT)

    We address the problem of existence of unbiased constrained parameter
    estimators. We show that if the constrained set of parameters is compact and
    the hypothesized distributions are absolutely continuous with respect to one
    another, then there exists no unbiased estimator. Weaker conditions for the
    absence of unbiased constrained estimators are also specified. We provide
    several examples which demonstrate the utility of these conditions.

    Signal acquisition via polarization modulation in single photon sources

    Mark D. McDonnell, Adrian P. Flitney
    Comments: 7 pages, 2 figures, accepted by Physical Review E. This version adds a reference
    Journal-ref: Physical Review E 80, 060102(R) (2009)
    Subjects: Quantum Physics (quant-ph); Information Theory (cs.IT)

    A simple model system is introduced for demonstrating how a single photon
    source might be used to transduce classical analog information. The theoretical
    scheme results in measurements of analog source samples that are (i) quantized
    in the sense of analog-to-digital conversion and (ii) corrupted by random noise
    that is solely due to the quantum uncertainty in detecting the polarization
    state of each photon. This noise is unavoidable if more than one bit per sample
    is to be transmitted, and we show how it may be exploited in a manner inspired
    by suprathreshold stochastic resonance. The system is analyzed information
    theoretically, as it can be modeled as a noisy optical communication channel,
    although unlike classical Poisson channels, the detector’s photon statistics
    are binomial. Previous results on binomial channels are adapted to demonstrate
    numerically that the classical information capacity, and thus the accuracy of
    the transduction, increases logarithmically with the square root of the number
    of photons, N. Although the capacity is shown to be reduced when an additional
    detector nonideality is present, the logarithmic increase with N remains.




沪ICP备19023445号-2号
友情链接