February 1, 2020

3345 words 16 mins read

Paper Group AWR 332

Single Deep Counterfactual Regret Minimization. A Convolutional Neural Network for the Automatic Diagnosis of Collagen VI related Muscular Dystrophies. Learning Mixtures of Separable Dictionaries for Tensor Data: Analysis and Algorithms. A Comprehensive Survey on Transfer Learning. SmartBullets: A Cloud-Assisted Bullet Screen Filter based on Deep L …

Single Deep Counterfactual Regret Minimization


Title	Single Deep Counterfactual Regret Minimization
Authors	Eric Steinberger
Abstract	Counterfactual Regret Minimization (CFR) is the most successful algorithm for finding approximate Nash equilibria in imperfect information games. However, CFR’s reliance on full game-tree traversals limits its scalability. For this reason, the game’s state- and action-space is often abstracted (i.e. simplified) for CFR, and the resulting strategy is then translated back to the full game, which requires extensive expert-knowledge and often converges to highly exploitable policies. A recently proposed method, Deep CFR, applies deep learning directly to CFR, allowing the agent to intrinsically abstract and generalize over the state-space from samples, without requiring expert knowledge. In this paper, we introduce Single Deep CFR (SD-CFR), a simplified variant of Deep CFR that has a lower overall approximation error by avoiding the training of an average strategy network. We show that SD-CFR is more attractive from a theoretical perspective and empirically outperforms Deep CFR with respect to exploitability and one-on-one play in poker.
Tasks
Published	2019-01-22
URL	https://arxiv.org/abs/1901.07621v4
PDF	https://arxiv.org/pdf/1901.07621v4.pdf
PWC	https://paperswithcode.com/paper/single-deep-counterfactual-regret
Repo	https://github.com/TopologicLogic/CFRM-ES-CFU-XGBoost
Framework	none


Title	A Convolutional Neural Network for the Automatic Diagnosis of Collagen VI related Muscular Dystrophies
Authors	Adrián Bazaga, Mònica Roldán, Carmen Badosa, Cecilia Jiménez-Mallebrera, Josep M. Porta
Abstract	The development of machine learning systems for the diagnosis of rare diseases is challenging mainly due the lack of data to study them. Despite this challenge, this paper proposes a system for the Computer Aided Diagnosis (CAD) of low-prevalence, congenital muscular dystrophies from confocal microscopy images. The proposed CAD system relies on a Convolutional Neural Network (CNN) which performs an independent classification for non-overlapping patches tiling the input image, and generates an overall decision summarizing the individual decisions for the patches on the query image. This decision scheme points to the possibly problematic areas in the input images and provides a global quantitative evaluation of the state of the patients, which is fundamental for diagnosis and to monitor the efficiency of therapies.
Tasks
Published	2019-01-30
URL	http://arxiv.org/abs/1901.11074v1
PDF	http://arxiv.org/pdf/1901.11074v1.pdf
PWC	https://paperswithcode.com/paper/a-convolutional-neural-network-for-the
Repo	https://github.com/AdrianBZG/Muscular-Dystrophy-Diagnosis
Framework	none

Learning Mixtures of Separable Dictionaries for Tensor Data: Analysis and Algorithms


Title	Learning Mixtures of Separable Dictionaries for Tensor Data: Analysis and Algorithms
Authors	Mohsen Ghassemi, Zahra Shakeri, Anand D. Sarwate, Waheed U. Bajwa
Abstract	This work addresses the problem of learning sparse representations of tensor data using structured dictionary learning. It proposes learning a mixture of separable dictionaries to better capture the structure of tensor data by generalizing the separable dictionary learning model. Two different approaches for learning mixture of separable dictionaries are explored and sufficient conditions for local identifiability of the underlying dictionary are derived in each case. Moreover, computational algorithms are developed to solve the problem of learning mixture of separable dictionaries in both batch and online settings. Numerical experiments are used to show the usefulness of the proposed model and the efficacy of the developed algorithms.
Tasks	Dictionary Learning
Published	2019-03-22
URL	http://arxiv.org/abs/1903.09284v1
PDF	http://arxiv.org/pdf/1903.09284v1.pdf
PWC	https://paperswithcode.com/paper/learning-mixtures-of-separable-dictionaries
Repo	https://github.com/MohsenGhassemi/STARK
Framework	none

A Comprehensive Survey on Transfer Learning


Title	A Comprehensive Survey on Transfer Learning
Authors	Fuzhen Zhuang, Zhiyuan Qi, Keyu Duan, Dongbo Xi, Yongchun Zhu, Hengshu Zhu, Hui Xiong, Qing He
Abstract	Transfer learning aims at improving the performance of target learners on target domains by transferring the knowledge contained in different but related source domains. In this way, the dependence on a large number of target domain data can be reduced for constructing target learners. Due to the wide application prospects, transfer learning has become a popular and promising area in machine learning. Although there are already some valuable and impressive surveys on transfer learning, these surveys introduce approaches in a relatively isolated way and lack the recent advances in transfer learning. As the rapid expansion of the transfer learning area, it is both necessary and challenging to comprehensively review the relevant studies. This survey attempts to connect and systematize the existing transfer learning researches, as well as to summarize and interpret the mechanisms and the strategies in a comprehensive way, which may help readers have a better understanding of the current research status and ideas. Different from previous surveys, this survey paper reviews over forty representative transfer learning approaches from the perspectives of data and model. The applications of transfer learning are also briefly introduced. In order to show the performance of different transfer learning models, twenty representative transfer learning models are used for experiments. The models are performed on three different datasets, i.e., Amazon Reviews, Reuters-21578, and Office-31. And the experimental results demonstrate the importance of selecting appropriate transfer learning models for different applications in practice.
Tasks	Transfer Learning
Published	2019-11-07
URL	https://arxiv.org/abs/1911.02685v2
PDF	https://arxiv.org/pdf/1911.02685v2.pdf
PWC	https://paperswithcode.com/paper/a-comprehensive-survey-on-transfer-learning
Repo	https://github.com/FuzhenZhuang/Transfer-Learning-Toolkit
Framework	none

SmartBullets: A Cloud-Assisted Bullet Screen Filter based on Deep Learning


Title	SmartBullets: A Cloud-Assisted Bullet Screen Filter based on Deep Learning
Authors	Haoran Niu, Jiangnan Li, Yu Zhao
Abstract	Bullet-screen is a technique that enables the website users to send real-time comment `bullet’ cross the screen. Compared with the traditional review of a video, bullet-screen provides new features of feeling expression to video watching and more iterations between video viewers. However, since all the comments from the viewers are shown on the screen publicly and simultaneously, some low-quality bullets will reduce the watching enjoyment of the users. Although the bullet-screen video websites have provided filter functions based on regular expression, bad bullets can still easily pass the filter through making a small modification. In this paper, we present SmartBullets, a user-centered bullet-screen filter based on deep learning techniques. A convolutional neural network is trained as the classifier to determine whether a bullet need to be removed according to its quality. Moreover, to increase the scalability of the filter, we employ a cloud-assisted framework by developing a backend cloud server and a front-end browser extension. The evaluation of 40 volunteers shows that SmartBullets can effectively remove the low-quality bullets and improve the overall watching experience of viewers. \|
Tasks
Published	2019-05-15
URL	https://arxiv.org/abs/1905.05925v1
PDF	https://arxiv.org/pdf/1905.05925v1.pdf
PWC	https://paperswithcode.com/paper/smartbullets-a-cloud-assisted-bullet-screen
Repo	https://github.com/hniu1/Smart-Bullet
Framework	tf

Information-geometric optimization with natural selection


Title	Information-geometric optimization with natural selection
Authors	Jakub Otwinowski, Colin LaMont
Abstract	Evolutionary algorithms, inspired by natural evolution, aim to optimize difficult objective functions without computing derivatives. Here we detail the relationship between population genetics and evolutionary optimization and formulate a new evolutionary algorithm. Optimization of a continuous objective function is analogous to searching for high fitness phenotypes on a fitness landscape. We summarize how natural selection moves a population along the non-euclidean gradient that is induced by the population on the fitness landscape (the natural gradient). Under normal approximations common in quantitative genetics, we show how selection is related to Newton’s method in optimization. We find that intermediate selection is most informative of the fitness landscape. We describe the generation of new phenotypes and introduce an operator that recombines the whole population to generate variants that preserve normal statistics. Finally, we introduce a proof-of-principle algorithm that combines natural selection, our recombination operator, and an adaptive method to increase selection. Our algorithm is similar to covariance matrix adaptation and natural evolutionary strategies in optimization, and has similar performance. The algorithm is extremely simple in implementation with no matrix inversion or factorization, does not require storing a covariance matrix, and may form the basis of more general model-based optimization algorithms with natural gradient updates.
Tasks
Published	2019-12-06
URL	https://arxiv.org/abs/1912.03395v2
PDF	https://arxiv.org/pdf/1912.03395v2.pdf
PWC	https://paperswithcode.com/paper/quantitative-genetic-algorithms
Repo	https://github.com/jotwin/qga
Framework	none

Transfer Learning Robustness in Multi-Class Categorization by Fine-Tuning Pre-Trained Contextualized Language Models


Title	Transfer Learning Robustness in Multi-Class Categorization by Fine-Tuning Pre-Trained Contextualized Language Models
Authors	Xinyi Liu, Artit Wangperawong
Abstract	This study compares the effectiveness and robustness of multi-class categorization of Amazon product data using transfer learning on pre-trained contextualized language models. Specifically, we fine-tuned BERT and XLNet, two bidirectional models that have achieved state-of-the-art performance on many natural language tasks and benchmarks, including text classification. While existing classification studies and benchmarks focus on binary targets, with the exception of ordinal ranking tasks, here we examine the robustness of such models as the number of classes grows from 1 to 20. Our experiments demonstrate an approximately linear decrease in performance metrics (i.e., precision, recall, $F_1$ score, and accuracy) with the number of class labels. BERT consistently outperforms XLNet using identical hyperparameters on the entire range of class label quantities for categorizing products based on their textual descriptions. BERT is also more affordable than XLNet in terms of the computational cost (i.e., time and memory) required for training. In all cases studied, the performance degradation rates were estimated to be 1% per additional class label.
Tasks	Text Classification, Transfer Learning
Published	2019-09-08
URL	https://arxiv.org/abs/1909.03564v2
PDF	https://arxiv.org/pdf/1909.03564v2.pdf
PWC	https://paperswithcode.com/paper/transfer-learning-robustness-in-multi-class
Repo	https://github.com/artitw/text2class
Framework	tf

Neural networks grown and self-organized by noise


Title	Neural networks grown and self-organized by noise
Authors	Guruprasad Raghavan, Matt Thomson
Abstract	Living neural networks emerge through a process of growth and self-organization that begins with a single cell and results in a brain, an organized and functional computational device. Artificial neural networks, however, rely on human-designed, hand-programmed architectures for their remarkable performance. Can we develop artificial computational devices that can grow and self-organize without human intervention? In this paper, we propose a biologically inspired developmental algorithm that can ‘grow’ a functional, layered neural network from a single initial cell. The algorithm organizes inter-layer connections to construct a convolutional pooling layer, a key constituent of convolutional neural networks (CNN’s). Our approach is inspired by the mechanisms employed by the early visual system to wire the retina to the lateral geniculate nucleus (LGN), days before animals open their eyes. The key ingredients for robust self-organization are an emergent spontaneous spatiotemporal activity wave in the first layer and a local learning rule in the second layer that ‘learns’ the underlying activity pattern in the first layer. The algorithm is adaptable to a wide-range of input-layer geometries, robust to malfunctioning units in the first layer, and so can be used to successfully grow and self-organize pooling architectures of different pool-sizes and shapes. The algorithm provides a primitive procedure for constructing layered neural networks through growth and self-organization. Broadly, our work shows that biologically inspired developmental algorithms can be applied to autonomously grow functional ‘brains’ in-silico.
Tasks
Published	2019-06-03
URL	https://arxiv.org/abs/1906.01039v1
PDF	https://arxiv.org/pdf/1906.01039v1.pdf
PWC	https://paperswithcode.com/paper/neural-networks-grown-and-self-organized-by
Repo	https://github.com/thomsonlab/GrowingBrains
Framework	none

Squeezed Very Deep Convolutional Neural Networks for Text Classification


Title	Squeezed Very Deep Convolutional Neural Networks for Text Classification
Authors	Andréa B. Duque, Luã Lázaro J. Santos, David Macêdo, Cleber Zanchettin
Abstract	Most of the research in convolutional neural networks has focused on increasing network depth to improve accuracy, resulting in a massive number of parameters which restricts the trained network to platforms with memory and processing constraints. We propose to modify the structure of the Very Deep Convolutional Neural Networks (VDCNN) model to fit mobile platforms constraints and keep performance. In this paper, we evaluate the impact of Temporal Depthwise Separable Convolutions and Global Average Pooling in the network parameters, storage size, and latency. The squeezed model (SVDCNN) is between 10x and 20x smaller, depending on the network depth, maintaining a maximum size of 6MB. Regarding accuracy, the network experiences a loss between 0.4% and 1.3% and obtains lower latencies compared to the baseline model.
Tasks	Sentiment Analysis, Text Classification
Published	2019-01-28
URL	http://arxiv.org/abs/1901.09821v1
PDF	http://arxiv.org/pdf/1901.09821v1.pdf
PWC	https://paperswithcode.com/paper/squeezed-very-deep-convolutional-neural
Repo	https://github.com/lazarotm/SVDCNN
Framework	pytorch

Temporal Self-Attention Network for Medical Concept Embedding


Title	Temporal Self-Attention Network for Medical Concept Embedding
Authors	Xueping Peng, Guodong Long, Tao Shen, Sen Wang, Jing Jiang, Michael Blumenstein
Abstract	In longitudinal electronic health records (EHRs), the event records of a patient are distributed over a long period of time and the temporal relations between the events reflect sufficient domain knowledge to benefit prediction tasks such as the rate of inpatient mortality. Medical concept embedding as a feature extraction method that transforms a set of medical concepts with a specific time stamp into a vector, which will be fed into a supervised learning algorithm. The quality of the embedding significantly determines the learning performance over the medical data. In this paper, we propose a medical concept embedding method based on applying a self-attention mechanism to represent each medical concept. We propose a novel attention mechanism which captures the contextual information and temporal relationships between medical concepts. A light-weight neural net, “Temporal Self-Attention Network (TeSAN)", is then proposed to learn medical concept embedding based solely on the proposed attention mechanism. To test the effectiveness of our proposed methods, we have conducted clustering and prediction tasks on two public EHRs datasets comparing TeSAN against five state-of-the-art embedding methods. The experimental results demonstrate that the proposed TeSAN model is superior to all the compared methods. To the best of our knowledge, this work is the first to exploit temporal self-attentive relations between medical events.
Tasks
Published	2019-09-15
URL	https://arxiv.org/abs/1909.06886v1
PDF	https://arxiv.org/pdf/1909.06886v1.pdf
PWC	https://paperswithcode.com/paper/temporal-self-attention-network-for-medical
Repo	https://github.com/cyrilmaxwell/MedicalAIPaperList
Framework	none

Robust and Subject-Independent Driving Manoeuvre Anticipation through Domain-Adversarial Recurrent Neural Networks


Title	Robust and Subject-Independent Driving Manoeuvre Anticipation through Domain-Adversarial Recurrent Neural Networks
Authors	Michele Tonutti, Emanuele Ruffaldi, Alessandro Cattaneo, Carlo Alberto Avizzano
Abstract	Through deep learning and computer vision techniques, driving manoeuvres can be predicted accurately a few seconds in advance. Even though adapting a learned model to new drivers and different vehicles is key for robust driver-assistance systems, this problem has received little attention so far. This work proposes to tackle this challenge through domain adaptation, a technique closely related to transfer learning. A proof of concept for the application of a Domain-Adversarial Recurrent Neural Network (DA-RNN) to multi-modal time series driving data is presented, in which domain-invariant features are learned by maximizing the loss of an auxiliary domain classifier. Our implementation is evaluated using a leave-one-driver-out approach on individual drivers from the Brain4Cars dataset, as well as using a new dataset acquired through driving simulations, yielding an average increase in performance of 30% and 114% respectively compared to no adaptation. We also show the importance of fine-tuning sections of the network to optimise the extraction of domain-independent features. The results demonstrate the applicability of the approach to driver-assistance systems as well as training and simulation environments.
Tasks	Domain Adaptation, Time Series, Transfer Learning
Published	2019-02-26
URL	http://arxiv.org/abs/1902.09820v1
PDF	http://arxiv.org/pdf/1902.09820v1.pdf
PWC	https://paperswithcode.com/paper/robust-and-subject-independent-driving
Repo	https://github.com/michetonu/DA-RNN_manoeuver_anticipation
Framework	tf

Snoopy: Sniffing Your Smartwatch Passwords via Deep Sequence Learning


Title	Snoopy: Sniffing Your Smartwatch Passwords via Deep Sequence Learning
Authors	Chris Xiaoxuan Lu, Bowen Du, Hongkai Wen, Sen Wang, Andrew Markham, Ivan Martinovic, Yiran Shen, Niki Trigoni
Abstract	Demand for smartwatches has taken off in recent years with new models which can run independently from smartphones and provide more useful features, becoming first-class mobile platforms. One can access online banking or even make payments on a smartwatch without a paired phone. This makes smartwatches more attractive and vulnerable to malicious attacks, which to date have been largely overlooked. In this paper, we demonstrate Snoopy, a password extraction and inference system which is able to accurately infer passwords entered on Android/Apple watches within 20 attempts, just by eavesdropping on motion sensors. Snoopy uses a uniform framework to extract the segments of motion data when passwords are entered, and uses novel deep neural networks to infer the actual passwords. We evaluate the proposed Snoopy system in the real-world with data from 362 participants and show that our system offers a 3-fold improvement in the accuracy of inferring passwords compared to the state-of-the-art, without consuming excessive energy or computational resources. We also show that Snoopy is very resilient to user and device heterogeneity: it can be trained on crowd-sourced motion data (e.g. via Amazon Mechanical Turk), and then used to attack passwords from a new user, even if they are wearing a different model. This paper shows that, in the wrong hands, Snoopy can potentially cause serious leaks of sensitive information. By raising awareness, we invite the community and manufacturers to revisit the risks of continuous motion sensing on smart wearable devices.
Tasks
Published	2019-12-10
URL	https://arxiv.org/abs/1912.04836v2
PDF	https://arxiv.org/pdf/1912.04836v2.pdf
PWC	https://paperswithcode.com/paper/snoopy-sniffing-your-smartwatch-passwords-via
Repo	https://github.com/ChristopherLu/snoopy
Framework	tf

Teacher algorithms for curriculum learning of Deep RL in continuously parameterized environments


Title	Teacher algorithms for curriculum learning of Deep RL in continuously parameterized environments
Authors	Rémy Portelas, Cédric Colas, Katja Hofmann, Pierre-Yves Oudeyer
Abstract	We consider the problem of how a teacher algorithm can enable an unknown Deep Reinforcement Learning (DRL) student to become good at a skill over a wide range of diverse environments. To do so, we study how a teacher algorithm can learn to generate a learning curriculum, whereby it sequentially samples parameters controlling a stochastic procedural generation of environments. Because it does not initially know the capacities of its student, a key challenge for the teacher is to discover which environments are easy, difficult or unlearnable, and in what order to propose them to maximize the efficiency of learning over the learnable ones. To achieve this, this problem is transformed into a surrogate continuous bandit problem where the teacher samples environments in order to maximize absolute learning progress of its student. We present a new algorithm modeling absolute learning progress with Gaussian mixture models (ALP-GMM). We also adapt existing algorithms and provide a complete study in the context of DRL. Using parameterized variants of the BipedalWalker environment, we study their efficiency to personalize a learning curriculum for different learners (embodiments), their robustness to the ratio of learnable/unlearnable environments, and their scalability to non-linear and high-dimensional parameter spaces. Videos and code are available at https://github.com/flowersteam/teachDeepRL.
Tasks
Published	2019-10-16
URL	https://arxiv.org/abs/1910.07224v1
PDF	https://arxiv.org/pdf/1910.07224v1.pdf
PWC	https://paperswithcode.com/paper/teacher-algorithms-for-curriculum-learning-of
Repo	https://github.com/flowersteam/teachDeepRL
Framework	tf

Google Research Football: A Novel Reinforcement Learning Environment


Title	Google Research Football: A Novel Reinforcement Learning Environment
Authors	Karol Kurach, Anton Raichuk, Piotr Stańczyk, Michał Zając, Olivier Bachem, Lasse Espeholt, Carlos Riquelme, Damien Vincent, Marcin Michalski, Olivier Bousquet, Sylvain Gelly
Abstract	Recent progress in the field of reinforcement learning has been accelerated by virtual learning environments such as video games, where novel algorithms and ideas can be quickly tested in a safe and reproducible manner. We introduce the Google Research Football Environment, a new reinforcement learning environment where agents are trained to play football in an advanced, physics-based 3D simulator. The resulting environment is challenging, easy to use and customize, and it is available under a permissive open-source license. In addition, it provides support for multiplayer and multi-agent experiments. We propose three full-game scenarios of varying difficulty with the Football Benchmarks and report baseline results for three commonly used reinforcement algorithms (IMPALA, PPO, and Ape-X DQN). We also provide a diverse set of simpler scenarios with the Football Academy and showcase several promising research directions.
Tasks	Game of Football
Published	2019-07-25
URL	https://arxiv.org/abs/1907.11180v1
PDF	https://arxiv.org/pdf/1907.11180v1.pdf
PWC	https://paperswithcode.com/paper/google-research-football-a-novel
Repo	https://github.com/google-research/football
Framework	tf

Towards Efficient Training for Neural Network Quantization


Title	Towards Efficient Training for Neural Network Quantization
Authors	Qing Jin, Linjie Yang, Zhenyu Liao
Abstract	Quantization reduces computation costs of neural networks but suffers from performance degeneration. Is this accuracy drop due to the reduced capacity, or inefficient training during the quantization procedure? After looking into the gradient propagation process of neural networks by viewing the weights and intermediate activations as random variables, we discover two critical rules for efficient training. Recent quantization approaches violates the two rules and results in degenerated convergence. To deal with this problem, we propose a simple yet effective technique, named scale-adjusted training (SAT), to comply with the discovered rules and facilitates efficient training. We also analyze the quantization error introduced in calculating the gradient in the popular parameterized clipping activation (PACT) technique. Through SAT together with gradient-calibrated PACT, quantized models obtain comparable or even better performance than their full-precision counterparts, achieving state-of-the-art accuracy with consistent improvement over previous quantization methods on a wide spectrum of models including MobileNet-V1/V2 and PreResNet-50.
Tasks	Quantization
Published	2019-12-21
URL	https://arxiv.org/abs/1912.10207v1
PDF	https://arxiv.org/pdf/1912.10207v1.pdf
PWC	https://paperswithcode.com/paper/towards-efficient-training-for-neural-network
Repo	https://github.com/jakc4103/scale-adjusted-training
Framework	pytorch