October 19, 2019

3110 words 15 mins read

Paper Group ANR 394

Best Arm Identification for Contaminated Bandits. Synthetic Perfusion Maps: Imaging Perfusion Deficits in DSC-MRI with Deep Learning. Assessing the Utility of Weather Data for Photovoltaic Power Prediction. Efficient Two-Step Adversarial Defense for Deep Neural Networks. The Voice Conversion Challenge 2018: Promoting Development of Parallel and Non …

Best Arm Identification for Contaminated Bandits


Title	Best Arm Identification for Contaminated Bandits
Authors	Jason Altschuler, Victor-Emmanuel Brunel, Alan Malek
Abstract	This paper studies active learning in the context of robust statistics. Specifically, we propose a variant of the Best Arm Identification problem for \emph{contaminated bandits}, where each arm pull has probability $\varepsilon$ of generating a sample from an arbitrary contamination distribution instead of the true underlying distribution. The goal is to identify the best (or approximately best) true distribution with high probability, with a secondary goal of providing guarantees on the quality of this distribution. The primary challenge of the contaminated bandit setting is that the true distributions are only partially identifiable, even with infinite samples. To address this, we develop tight, non-asymptotic sample complexity bounds for high-probability estimation of the first two robust moments (median and median absolute deviation) from contaminated samples. These concentration inequalities are the main technical contributions of the paper and may be of independent interest. Using these results, we adapt several classical Best Arm Identification algorithms to the contaminated bandit setting and derive sample complexity upper bounds for our problem. Finally, we provide matching information-theoretic lower bounds on the sample complexity (up to a small logarithmic factor).
Tasks	Active Learning
Published	2018-02-26
URL	https://arxiv.org/abs/1802.09514v5
PDF	https://arxiv.org/pdf/1802.09514v5.pdf
PWC	https://paperswithcode.com/paper/best-arm-identification-for-contaminated
Repo
Framework

Synthetic Perfusion Maps: Imaging Perfusion Deficits in DSC-MRI with Deep Learning


Title	Synthetic Perfusion Maps: Imaging Perfusion Deficits in DSC-MRI with Deep Learning
Authors	Andreas Hess, Raphael Meier, Johannes Kaesmacher, Simon Jung, Fabien Scalzo, David Liebeskind, Roland Wiest, Richard McKinley
Abstract	In this work, we present a novel convolutional neural net- work based method for perfusion map generation in dynamic suscepti- bility contrast-enhanced perfusion imaging. The proposed architecture is trained end-to-end and solely relies on raw perfusion data for inference. We used a dataset of 151 acute ischemic stroke cases for evaluation. Our method generates perfusion maps that are comparable to the target maps used for clinical routine, while being model-free, fast, and less noisy.
Tasks
Published	2018-06-11
URL	http://arxiv.org/abs/1806.03848v1
PDF	http://arxiv.org/pdf/1806.03848v1.pdf
PWC	https://paperswithcode.com/paper/synthetic-perfusion-maps-imaging-perfusion
Repo
Framework

Assessing the Utility of Weather Data for Photovoltaic Power Prediction


Title	Assessing the Utility of Weather Data for Photovoltaic Power Prediction
Authors	Reza Zafarani, Sara Eftekharnejad, Urvi Patel
Abstract	Photovoltaic systems have been widely deployed in recent times to meet the increased electricity demand as an environmental-friendly energy source. The major challenge for integrating photovoltaic systems in power systems is the unpredictability of the solar power generated. In this paper, we analyze the impact of having access to weather information for solar power generation prediction and find weather information that can help best predict photovoltaic power.
Tasks
Published	2018-02-12
URL	http://arxiv.org/abs/1802.03913v1
PDF	http://arxiv.org/pdf/1802.03913v1.pdf
PWC	https://paperswithcode.com/paper/assessing-the-utility-of-weather-data-for
Repo
Framework

Efficient Two-Step Adversarial Defense for Deep Neural Networks


Title	Efficient Two-Step Adversarial Defense for Deep Neural Networks
Authors	Ting-Jui Chang, Yukun He, Peng Li
Abstract	In recent years, deep neural networks have demonstrated outstanding performance in many machine learning tasks. However, researchers have discovered that these state-of-the-art models are vulnerable to adversarial examples: legitimate examples added by small perturbations which are unnoticeable to human eyes. Adversarial training, which augments the training data with adversarial examples during the training process, is a well known defense to improve the robustness of the model against adversarial attacks. However, this robustness is only effective to the same attack method used for adversarial training. Madry et al.(2017) suggest that effectiveness of iterative multi-step adversarial attacks and particularly that projected gradient descent (PGD) may be considered the universal first order adversary and applying the adversarial training with PGD implies resistance against many other first order attacks. However, the computational cost of the adversarial training with PGD and other multi-step adversarial examples is much higher than that of the adversarial training with other simpler attack techniques. In this paper, we show how strong adversarial examples can be generated only at a cost similar to that of two runs of the fast gradient sign method (FGSM), allowing defense against adversarial attacks with a robustness level comparable to that of the adversarial training with multi-step adversarial examples. We empirically demonstrate the effectiveness of the proposed two-step defense approach against different attack methods and its improvements over existing defense strategies.
Tasks	Adversarial Defense
Published	2018-10-08
URL	http://arxiv.org/abs/1810.03739v1
PDF	http://arxiv.org/pdf/1810.03739v1.pdf
PWC	https://paperswithcode.com/paper/efficient-two-step-adversarial-defense-for
Repo
Framework

The Voice Conversion Challenge 2018: Promoting Development of Parallel and Nonparallel Methods


Title	The Voice Conversion Challenge 2018: Promoting Development of Parallel and Nonparallel Methods
Authors	Jaime Lorenzo-Trueba, Junichi Yamagishi, Tomoki Toda, Daisuke Saito, Fernando Villavicencio, Tomi Kinnunen, Zhenhua Ling
Abstract	We present the Voice Conversion Challenge 2018, designed as a follow up to the 2016 edition with the aim of providing a common framework for evaluating and comparing different state-of-the-art voice conversion (VC) systems. The objective of the challenge was to perform speaker conversion (i.e. transform the vocal identity) of a source speaker to a target speaker while maintaining linguistic information. As an update to the previous challenge, we considered both parallel and non-parallel data to form the Hub and Spoke tasks, respectively. A total of 23 teams from around the world submitted their systems, 11 of them additionally participated in the optional Spoke task. A large-scale crowdsourced perceptual evaluation was then carried out to rate the submitted converted speech in terms of naturalness and similarity to the target speaker identity. In this paper, we present a brief summary of the state-of-the-art techniques for VC, followed by a detailed explanation of the challenge tasks and the results that were obtained.
Tasks	Voice Conversion
Published	2018-04-12
URL	http://arxiv.org/abs/1804.04262v1
PDF	http://arxiv.org/pdf/1804.04262v1.pdf
PWC	https://paperswithcode.com/paper/the-voice-conversion-challenge-2018-promoting
Repo
Framework

Learning Analytics in Massive Open Online Courses


Title	Learning Analytics in Massive Open Online Courses
Authors	Mohammad Khalil
Abstract	Educational technology has obtained great importance over the last fifteen years. At present, the umbrella of educational technology incorporates multitudes of engaging online environments and fields. Learning analytics and Massive Open Online Courses (MOOCs) are two of the most relevant emerging topics in this domain. Since they are open to everyone at no cost, MOOCs excel in attracting numerous participants that can reach hundreds and hundreds of thousands. Experts from different disciplines have shown significant interest in MOOCs as the phenomenon has rapidly grown. In fact, MOOCs have been proven to scale education in disparate areas. Their benefits are crystallized in the improvement of educational outcomes, reduction of costs and accessibility expansion. Due to their unusual massiveness, the large datasets of MOOC platforms require advanced tools and methodologies for further examination. The key importance of learning analytics is reflected here. MOOCs offer diverse challenges and practices for learning analytics to tackle. In view of that, this thesis combines both fields in order to investigate further steps in the learning analytics capabilities in MOOCs. The primary research of this dissertation focuses on the integration of learning analytics in MOOCs, and thereafter looks into examining students’ behavior on one side and bridging MOOC issues on the other side. The research was done on the Austrian iMooX xMOOC platform. We followed the prototyping and case studies research methodology to carry out the research questions of this dissertation. The main contributions incorporate designing a general learning analytics framework, learning analytics prototype, records of students’ behavior in nearly every MOOC’s variables (discussion forums, interactions in videos, self-assessment quizzes, login frequency), a cluster of student engagement…
Tasks
Published	2018-02-17
URL	http://arxiv.org/abs/1802.09344v1
PDF	http://arxiv.org/pdf/1802.09344v1.pdf
PWC	https://paperswithcode.com/paper/learning-analytics-in-massive-open-online
Repo
Framework

Model-Protected Multi-Task Learning


Title	Model-Protected Multi-Task Learning
Authors	Jian Liang, Ziqi Liu, Jiayu Zhou, Xiaoqian Jiang, Changshui Zhang, Fei Wang
Abstract	Multi-task learning (MTL) refers to the paradigm of learning multiple related tasks together. In contrast, in single-task learning (STL) each individual task is learned independently. MTL often leads to better trained models because they can leverage the commonalities among related tasks. However, because MTL algorithms can ``leak” information from different models across different tasks, MTL poses a potential security risk. Specifically, an adversary may participate in the MTL process through one task and thereby acquire the model information for another task. The previously proposed privacy-preserving MTL methods protect data instances rather than models, and some of them may underperform in comparison with STL methods. In this paper, we propose a privacy-preserving MTL framework to prevent information from each model leaking to other models based on a perturbation of the covariance matrix of the model matrix. We study two popular MTL approaches for instantiation, namely, learning the low-rank and group-sparse patterns of the model matrix. Our algorithms can be guaranteed not to underperform compared with STL methods. We build our methods based upon tools for differential privacy, and privacy guarantees, utility bounds are provided, and heterogeneous privacy budgets are considered. The experiments demonstrate that our algorithms outperform the baseline methods constructed by existing privacy-preserving MTL methods on the proposed model-protection problem. \|
Tasks	Multi-Task Learning
Published	2018-09-18
URL	https://arxiv.org/abs/1809.06546v3
PDF	https://arxiv.org/pdf/1809.06546v3.pdf
PWC	https://paperswithcode.com/paper/model-protected-multi-task-learning
Repo
Framework

Making \emph{ordinary least squares} linear classfiers more robust


Title	Making \emph{ordinary least squares} linear classfiers more robust
Authors	Babatunde M. Ayeni
Abstract	In the field of statistics and machine learning, the sums-of-squares, commonly referred to as \emph{ordinary least squares}, can be used as a convenient choice of cost function because of its many nice analytical properties, though not always the best choice. However, it has been long known that \emph{ordinary least squares} is not robust to outliers. Several attempts to resolve this problem led to the creation of alternative methods that, either did not fully resolved the \emph{outlier problem} or were computationally difficult. In this paper, we provide a very simple solution that can make \emph{ordinary least squares} less sensitive to outliers in data classification, by \emph{scaling the augmented input vector by its length}. We show some mathematical expositions of the \emph{outlier problem} using some approximations and geometrical techniques. We present numerical results to support the efficacy of our method.
Tasks
Published	2018-08-28
URL	http://arxiv.org/abs/1808.09222v1
PDF	http://arxiv.org/pdf/1808.09222v1.pdf
PWC	https://paperswithcode.com/paper/making-emphordinary-least-squares-linear
Repo
Framework

Rate Distortion For Model Compression: From Theory To Practice


Title	Rate Distortion For Model Compression: From Theory To Practice
Authors	Weihao Gao, Yu-Han Liu, Chong Wang, Sewoong Oh
Abstract	The enormous size of modern deep neural networks makes it challenging to deploy those models in memory and communication limited scenarios. Thus, compressing a trained model without a significant loss in performance has become an increasingly important task. Tremendous advances has been made recently, where the main technical building blocks are parameter pruning, parameter sharing (quantization), and low-rank factorization. In this paper, we propose principled approaches to improve upon the common heuristics used in those building blocks, namely pruning and quantization. We first study the fundamental limit for model compression via the rate distortion theory. We bring the rate distortion function from data compression to model compression to quantify this fundamental limit. We prove a lower bound for the rate distortion function and prove its achievability for linear models. Although this achievable compression scheme is intractable in practice, this analysis motivates a novel model compression framework. This framework provides a new objective function in model compression, which can be applied together with other classes of model compressor such as pruning or quantization. Theoretically, we prove that the proposed scheme is optimal for compressing one-hidden-layer ReLU neural networks. Empirically, we show that the proposed scheme improves upon the baseline in the compression-accuracy tradeoff.
Tasks	Model Compression, Quantization
Published	2018-10-09
URL	http://arxiv.org/abs/1810.06401v2
PDF	http://arxiv.org/pdf/1810.06401v2.pdf
PWC	https://paperswithcode.com/paper/rate-distortion-for-model-compression-from
Repo
Framework

Evolutionary Computation plus Dynamic Programming for the Bi-Objective Travelling Thief Problem


Title	Evolutionary Computation plus Dynamic Programming for the Bi-Objective Travelling Thief Problem
Authors	Junhua Wu, Sergey Polyakovskiy, Markus Wagner, Frank Neumann
Abstract	This research proposes a novel indicator-based hybrid evolutionary approach that combines approximate and exact algorithms. We apply it to a new bi-criteria formulation of the travelling thief problem, which is known to the Evolutionary Computation community as a benchmark multi-component optimisation problem that interconnects two classical NP-hard problems: the travelling salesman problem and the 0-1 knapsack problem. Our approach employs the exact dynamic programming algorithm for the underlying Packing-While-Travelling (PWT) problem as a subroutine within a bi-objective evolutionary algorithm. This design takes advantage of the data extracted from Pareto fronts generated by the dynamic program to achieve better solutions. Furthermore, we develop a number of novel indicators and selection mechanisms to strengthen synergy of the two algorithmic components of our approach. The results of computational experiments show that the approach is capable to outperform the state-of-the-art results for the single-objective case of the problem.
Tasks
Published	2018-02-07
URL	http://arxiv.org/abs/1802.02434v1
PDF	http://arxiv.org/pdf/1802.02434v1.pdf
PWC	https://paperswithcode.com/paper/evolutionary-computation-plus-dynamic
Repo
Framework

Design of iMacros-based Data Crawler and the Behavioral Analysis of Facebook Users


Title	Design of iMacros-based Data Crawler and the Behavioral Analysis of Facebook Users
Authors	Mudasir Ahmad Wani, Nancy Agarwal, Suraiya Jabin, Syed Zeeshan Hussai
Abstract	Obtaining the desired dataset is still a prime challenge faced by researchers while analyzing Online Social Network (OSN) sites. Application Programming Interfaces (APIs) provided by OSN service providers for retrieving data impose several unavoidable restrictions which make it difficult to get a desirable dataset. In this paper, we present an iMacros technology-based data crawler called IMcrawler, capable of collecting every piece of information which is accessible through a browser from the Facebook website within the legal framework which permits access to publicly shared user content on OSNs. The proposed crawler addresses most of the challenges allied with web data extraction approaches and most of the APIs provided by OSN service providers. Two broad sections have been extracted from Facebook user profiles, namely, Personal Information and Wall Activities. The present work is the first attempt towards providing the detailed description of crawler design for the Facebook website.
Tasks
Published	2018-02-18
URL	https://arxiv.org/abs/1802.09566v2
PDF	https://arxiv.org/pdf/1802.09566v2.pdf
PWC	https://paperswithcode.com/paper/design-of-imacros-based-data-crawler-and-the
Repo
Framework

How does Lipschitz Regularization Influence GAN Training?


Title	How does Lipschitz Regularization Influence GAN Training?
Authors	Yipeng Qin, Niloy Mitra, Peter Wonka
Abstract	Lipschitz regularization has shown great success in stabilizing GAN training. Despite the recent algorithmic progress (e.g., weight clipping [2], gradient penalty [5], spectral normalization [18]), the exact reason of its effectiveness is still not well understood. The common belief is that Lipschitz regularization works by keeping the weights of the discriminator network small and thereby bounding the neural network gradients. In this work, we discover an even more important function of Lipschitz regularization. It restricts the domain, range, and interval of attainable gradient values of loss functions, thereby preventing the backpropagation of vanishing or exploding gradients. By introducing the concept of domain scaling, we can decouple the effect of Lipschitz regularization on the discriminator network and the loss function. This enables us to show that discriminator networks with larger weights also perform well as long as the domain of the loss function is scaled down. Empirically, we verify our proposition on the MNIST, CIFAR10 and CelebA datasets.
Tasks
Published	2018-11-23
URL	http://arxiv.org/abs/1811.09567v2
PDF	http://arxiv.org/pdf/1811.09567v2.pdf
PWC	https://paperswithcode.com/paper/do-gan-loss-functions-really-matter
Repo
Framework

82 Treebanks, 34 Models: Universal Dependency Parsing with Multi-Treebank Models


Title	82 Treebanks, 34 Models: Universal Dependency Parsing with Multi-Treebank Models
Authors	Aaron Smith, Bernd Bohnet, Miryam de Lhoneux, Joakim Nivre, Yan Shao, Sara Stymne
Abstract	We present the Uppsala system for the CoNLL 2018 Shared Task on universal dependency parsing. Our system is a pipeline consisting of three components: the first performs joint word and sentence segmentation; the second predicts part-of- speech tags and morphological features; the third predicts dependency trees from words and tags. Instead of training a single parsing model for each treebank, we trained models with multiple treebanks for one language or closely related languages, greatly reducing the number of models. On the official test run, we ranked 7th of 27 teams for the LAS and MLAS metrics. Our system obtained the best scores overall for word segmentation, universal POS tagging, and morphological features.
Tasks	Dependency Parsing
Published	2018-09-06
URL	http://arxiv.org/abs/1809.02237v1
PDF	http://arxiv.org/pdf/1809.02237v1.pdf
PWC	https://paperswithcode.com/paper/82-treebanks-34-models-universal-dependency
Repo
Framework

Transferable Positive/Negative Speech Emotion Recognition via Class-wise Adversarial Domain Adaptation


Title	Transferable Positive/Negative Speech Emotion Recognition via Class-wise Adversarial Domain Adaptation
Authors	Hao Zhou, Ke Chen
Abstract	Speech emotion recognition plays an important role in building more intelligent and human-like agents. Due to the difficulty of collecting speech emotional data, an increasingly popular solution is leveraging a related and rich source corpus to help address the target corpus. However, domain shift between the corpora poses a serious challenge, making domain shift adaptation difficult to function even on the recognition of positive/negative emotions. In this work, we propose class-wise adversarial domain adaptation to address this challenge by reducing the shift for all classes between different corpora. Experiments on the well-known corpora EMODB and Aibo demonstrate that our method is effective even when only a very limited number of target labeled examples are provided.
Tasks	Domain Adaptation, Emotion Recognition, Speech Emotion Recognition
Published	2018-10-30
URL	http://arxiv.org/abs/1810.12782v2
PDF	http://arxiv.org/pdf/1810.12782v2.pdf
PWC	https://paperswithcode.com/paper/transferable-positivenegative-speech-emotion
Repo
Framework

Classification of Things in DBpedia using Deep Neural Networks


Title	Classification of Things in DBpedia using Deep Neural Networks
Authors	Rahul Parundekar
Abstract	The Semantic Web aims at representing knowledge about the real world at web scale - things, their attributes and relationships among them can be represented as nodes and edges in an inter-linked semantic graph. In the presence of noisy data, as is typical of data on the Semantic Web, a software Agent needs to be able to robustly infer one or more associated actionable classes for the individuals in order to act automatically on it. We model this problem as a multi-label classification task where we want to robustly identify types of the individuals in a semantic graph such as DBpedia, which we use as an exemplary dataset on the Semantic Web. Our approach first extracts multiple features for the individuals using random walks and then performs multi-label classification using fully-connected Neural Networks. Through systematic exploration and experimentation, we identify the effect of hyper-parameters of the feature extraction and the fully-connected Neural Network structure on the classification performance. Our final results show that our method performs better than state-of-the-art inferencing systems like SDtype and SLCN, from which we can conclude that random-walk-based feature extraction of individuals and their multi-label classification using Deep Neural Networks is a promising alternative to these systems for type classification of individuals on the Semantic Web. The main contribution of our work is to introduce a novel approach that allows us to use Deep Neural Networks to identify types of individuals in a noisy semantic graph by extracting features using random walks
Tasks	Multi-Label Classification
Published	2018-02-07
URL	http://arxiv.org/abs/1802.02528v1
PDF	http://arxiv.org/pdf/1802.02528v1.pdf
PWC	https://paperswithcode.com/paper/classification-of-things-in-dbpedia-using
Repo
Framework