January 27, 2020

3146 words 15 mins read

Paper Group ANR 1208

Continual Learning in Neural Networks. Automatic Detection of Cortical Arousals in Sleep and their Contribution to Daytime Sleepiness. EM-NET: Centerline-Aware Mitochondria Segmentation in EM Images via Hierarchical View-Ensemble Convolutional Network. Locally Connected Spiking Neural Networks for Unsupervised Feature Learning. Online Object Repres …

Continual Learning in Neural Networks


Title	Continual Learning in Neural Networks
Authors	Rahaf Aljundi
Abstract	Artificial neural networks have exceeded human-level performance in accomplishing several individual tasks (e.g. voice recognition, object recognition, and video games). However, such success remains modest compared to human intelligence that can learn and perform an unlimited number of tasks. Humans’ ability of learning and accumulating knowledge over their lifetime is an essential aspect of their intelligence. Continual machine learning aims at a higher level of machine intelligence through providing the artificial agents with the ability to learn online from a non-stationary and never-ending stream of data. A key component of such a never-ending learning process is to overcome the catastrophic forgetting of previously seen data, a problem that neural networks are well known to suffer from. The work described in this thesis has been dedicated to the investigation of continual learning and solutions to mitigate the forgetting phenomena in neural networks. To approach the continual learning problem, we first assume a task incremental setting where tasks are received one at a time and data from previous tasks are not stored. Since the task incremental setting can’t be assumed in all continual learning scenarios, we also study the more general online continual setting. We consider an infinite stream of data drawn from a non-stationary distribution with a supervisory or self-supervisory training signal. The proposed methods in this thesis have tackled important aspects of continual learning. They were evaluated on different benchmarks and over various learning sequences. Advances in the state of the art of continual learning have been shown and challenges for bringing continual learning into application were critically identified.
Tasks	Continual Learning, Object Recognition
Published	2019-10-07
URL	https://arxiv.org/abs/1910.02718v2
PDF	https://arxiv.org/pdf/1910.02718v2.pdf
PWC	https://paperswithcode.com/paper/continual-learning-in-neural-networks
Repo
Framework

Automatic Detection of Cortical Arousals in Sleep and their Contribution to Daytime Sleepiness


Title	Automatic Detection of Cortical Arousals in Sleep and their Contribution to Daytime Sleepiness
Authors	Andreas Brink-Kjaer, Alexander Neergaard Olesen, Paul E. Peppard, Katie L. Stone, Poul Jennum, Emmanuel Mignot, Helge B. D. Sorensen
Abstract	Cortical arousals are transient events of disturbed sleep that occur spontaneously or in response to stimuli such as apneic events. The gold standard for arousal detection in human polysomnographic recordings (PSGs) is manual annotation by expert human scorers, a method with significant interscorer variability. In this study, we developed an automated method, the Multimodal Arousal Detector (MAD), to detect arousals using deep learning methods. The MAD was trained on 2,889 PSGs to detect both cortical arousals and wakefulness in 1 second intervals. Furthermore, the relationship between MAD-predicted labels on PSGs and next day mean sleep latency (MSL) on a multiple sleep latency test (MSLT), a reflection of daytime sleepiness, was analyzed in 1447 MSLT instances in 873 subjects. In a dataset of 1,026 PSGs, the MAD achieved a F1 score of 0.76 for arousal detection, while wakefulness was predicted with an accuracy of 0.95. In 60 PSGs scored by multiple human expert technicians, the MAD significantly outperformed the average human scorer for arousal detection with a difference in F1 score of 0.09. After controlling for other known covariates, a doubling of the arousal index was associated with an average decrease in MSL of 40 seconds ($\beta$ = -0.67, p = 0.0075). The MAD outperformed the average human expert and the MAD-predicted arousals were shown to be significant predictors of MSL, which demonstrate clinical validity the MAD.
Tasks
Published	2019-03-15
URL	http://arxiv.org/abs/1906.01700v1
PDF	http://arxiv.org/pdf/1906.01700v1.pdf
PWC	https://paperswithcode.com/paper/190601700
Repo
Framework

EM-NET: Centerline-Aware Mitochondria Segmentation in EM Images via Hierarchical View-Ensemble Convolutional Network


Title	EM-NET: Centerline-Aware Mitochondria Segmentation in EM Images via Hierarchical View-Ensemble Convolutional Network
Authors	Zhimin Yuan, Jiajin Yi, Zhengrong Luo, Zhongdao Jia, Jialin Peng
Abstract	Although deep encoder-decoder networks have achieved astonishing performance for mitochondria segmentation from electron microscopy (EM) images, they still produce coarse segmentations with lots of discontinuities and false positives. Besides, the need for labor intensive annotations of large 3D dataset and huge memory overhead by 3D models are also major limitations. To address these problems, we introduce a multi-task network named EM-Net, which includes an auxiliary centerline detection task to account for shape information of mitochondria represented by centerline. Therefore, the centerline detection sub-network is able to enhance the accuracy and robustness of segmentation task, especially when only a small set of annotated data are available. To achieve a light-weight 3D network, we introduce a novel hierarchical view-ensemble convolution module to reduce number of parameters, and facilitate multi-view information aggregation.Validations on public benchmark showed state-of-the-art performance by EM-Net. Even with significantly reduced training data, our method still showed quite promising results.
Tasks
Published	2019-11-30
URL	https://arxiv.org/abs/1912.00201v3
PDF	https://arxiv.org/pdf/1912.00201v3.pdf
PWC	https://paperswithcode.com/paper/em-net-centerline-aware-mitochondria
Repo
Framework

Locally Connected Spiking Neural Networks for Unsupervised Feature Learning


Title	Locally Connected Spiking Neural Networks for Unsupervised Feature Learning
Authors	Daniel J. Saunders, Devdhar Patel, Hananel Hazan, Hava T. Siegelmann, Robert Kozma
Abstract	In recent years, Spiking Neural Networks (SNNs) have demonstrated great successes in completing various Machine Learning tasks. We introduce a method for learning image features by \textit{locally connected layers} in SNNs using spike-timing-dependent plasticity (STDP) rule. In our approach, sub-networks compete via competitive inhibitory interactions to learn features from different locations of the input space. These \textit{Locally-Connected SNNs} (LC-SNNs) manifest key topological features of the spatial interaction of biological neurons. We explore biologically inspired n-gram classification approach allowing parallel processing over various patches of the the image space. We report the classification accuracy of simple two-layer LC-SNNs on two image datasets, which match the state-of-art performance and are the first results to date. LC-SNNs have the advantage of fast convergence to a dataset representation, and they require fewer learnable parameters than other SNN approaches with unsupervised learning. Robustness tests demonstrate that LC-SNNs exhibit graceful degradation of performance despite the random deletion of large amounts of synapses and neurons.
Tasks
Published	2019-04-12
URL	http://arxiv.org/abs/1904.06269v1
PDF	http://arxiv.org/pdf/1904.06269v1.pdf
PWC	https://paperswithcode.com/paper/locally-connected-spiking-neural-networks-for
Repo
Framework

Online Object Representations with Contrastive Learning


Title	Online Object Representations with Contrastive Learning
Authors	Sören Pirk, Mohi Khansari, Yunfei Bai, Corey Lynch, Pierre Sermanet
Abstract	We propose a self-supervised approach for learning representations of objects from monocular videos and demonstrate it is particularly useful in situated settings such as robotics. The main contributions of this paper are: 1) a self-supervising objective trained with contrastive learning that can discover and disentangle object attributes from video without using any labels; 2) we leverage object self-supervision for online adaptation: the longer our online model looks at objects in a video, the lower the object identification error, while the offline baseline remains with a large fixed error; 3) to explore the possibilities of a system entirely free of human supervision, we let a robot collect its own data, train on this data with our self-supervise scheme, and then show the robot can point to objects similar to the one presented in front of it, demonstrating generalization of object attributes. An interesting and perhaps surprising finding of this approach is that given a limited set of objects, object correspondences will naturally emerge when using contrastive learning without requiring explicit positive pairs. Videos illustrating online object adaptation and robotic pointing are available at: https://online-objects.github.io/.
Tasks
Published	2019-06-10
URL	https://arxiv.org/abs/1906.04312v1
PDF	https://arxiv.org/pdf/1906.04312v1.pdf
PWC	https://paperswithcode.com/paper/online-object-representations-with
Repo
Framework

Exploratory Not Explanatory: Counterfactual Analysis of Saliency Maps for Deep Reinforcement Learning


Title	Exploratory Not Explanatory: Counterfactual Analysis of Saliency Maps for Deep Reinforcement Learning
Authors	Akanksha Atrey, Kaleigh Clary, David Jensen
Abstract	Saliency maps are frequently used to support explanations of the behavior of deep reinforcement learning (RL) agents. However, a review of how saliency maps are used in practice indicates that the derived explanations are often unfalsifiable and can be highly subjective. We introduce an empirical approach grounded in counterfactual reasoning to test the hypotheses generated from saliency maps and assess the degree to which they correspond to the semantics of RL environments. We use Atari games, a common benchmark for deep RL, to evaluate three types of saliency maps. Our results show the extent to which existing claims about Atari games can be evaluated and suggest that saliency maps are best viewed as an exploratory tool rather than an explanatory tool.
Tasks	Atari Games
Published	2019-12-09
URL	https://arxiv.org/abs/1912.05743v2
PDF	https://arxiv.org/pdf/1912.05743v2.pdf
PWC	https://paperswithcode.com/paper/exploratory-not-explanatory-counterfactual-1
Repo
Framework

Extracting deep local features to detect manipulated images of human faces


Title	Extracting deep local features to detect manipulated images of human faces
Authors	Michail Tarasiou, Stefanos Zafeiriou
Abstract	Recent developments in computer vision and machine learning have made it possible to create realistic manipulated videos of human faces, raising the issue of ensuring adequate protection against the malevolent effects unlocked by such capabilities. In this paper we propose local image features that are shared across manipulated regions are the key element for the automatic detection of manipulated face images. We also design a lightweight architecture with the correct structural biases for extracting such features and derive a multitask training scheme that consistently outperforms image class supervision alone. The trained networks achieve state-of-the-art results in the FaceForensics++ dataset using significantly reduced number of parameters and are shown to work well in detecting fully generated face images.
Tasks
Published	2019-11-29
URL	https://arxiv.org/abs/1911.13269v2
PDF	https://arxiv.org/pdf/1911.13269v2.pdf
PWC	https://paperswithcode.com/paper/using-fully-convolutional-neural-networks-to
Repo
Framework

Uncertainty-Aware Imitation Learning using Kernelized Movement Primitives


Title	Uncertainty-Aware Imitation Learning using Kernelized Movement Primitives
Authors	João Silvério, Yanlong Huang, Fares J. Abu-Dakka, Leonel Rozo, Darwin G. Caldwell
Abstract	During the past few years, probabilistic approaches to imitation learning have earned a relevant place in the literature. One of their most prominent features, in addition to extracting a mean trajectory from task demonstrations, is that they provide a variance estimation. The intuitive meaning of this variance, however, changes across different techniques, indicating either variability or uncertainty. In this paper we leverage kernelized movement primitives (KMP) to provide a new perspective on imitation learning by predicting variability, correlations and uncertainty about robot actions. This rich set of information is used in combination with optimal controller fusion to learn actions from data, with two main advantages: i) robots become safe when uncertain about their actions and ii) they are able to leverage partial demonstrations, given as elementary sub-tasks, to optimally perform a higher level, more complex task. We showcase our approach in a painting task, where a human user and a KUKA robot collaborate to paint a wooden board. The task is divided into two sub-tasks and we show that using our approach the robot becomes compliant (hence safe) outside the training regions and executes the two sub-tasks with optimal gains.
Tasks	Imitation Learning
Published	2019-03-05
URL	https://arxiv.org/abs/1903.02114v3
PDF	https://arxiv.org/pdf/1903.02114v3.pdf
PWC	https://paperswithcode.com/paper/uncertainty-aware-imitation-learning-using
Repo
Framework

Enhancing Time Series Momentum Strategies Using Deep Neural Networks


Title	Enhancing Time Series Momentum Strategies Using Deep Neural Networks
Authors	Bryan Lim, Stefan Zohren, Stephen Roberts
Abstract	While time series momentum is a well-studied phenomenon in finance, common strategies require the explicit definition of both a trend estimator and a position sizing rule. In this paper, we introduce Deep Momentum Networks – a hybrid approach which injects deep learning based trading rules into the volatility scaling framework of time series momentum. The model also simultaneously learns both trend estimation and position sizing in a data-driven manner, with networks directly trained by optimising the Sharpe ratio of the signal. Backtesting on a portfolio of 88 continuous futures contracts, we demonstrate that the Sharpe-optimised LSTM improved traditional methods by more than two times in the absence of transactions costs, and continue outperforming when considering transaction costs up to 2-3 basis points. To account for more illiquid assets, we also propose a turnover regularisation term which trains the network to factor in costs at run-time.
Tasks	Time Series
Published	2019-04-09
URL	https://arxiv.org/abs/1904.04912v2
PDF	https://arxiv.org/pdf/1904.04912v2.pdf
PWC	https://paperswithcode.com/paper/enhancing-time-series-momentum-strategies
Repo
Framework

A system for the 2019 Sentiment, Emotion and Cognitive State Task of DARPAs LORELEI project


Title	A system for the 2019 Sentiment, Emotion and Cognitive State Task of DARPAs LORELEI project
Authors	Victor R Martinez, Anil Ramakrishna, Ming-Chang Chiu, Karan Singla, Shrikanth Narayanan
Abstract	During the course of a Humanitarian Assistance-Disaster Relief (HADR) crisis, that can happen anywhere in the world, real-time information is often posted online by the people in need of help which, in turn, can be used by different stakeholders involved with management of the crisis. Automated processing of such posts can considerably improve the effectiveness of such efforts; for example, understanding the aggregated emotion from affected populations in specific areas may help inform decision-makers on how to best allocate resources for an effective disaster response. However, these efforts may be severely limited by the availability of resources for the local language. The ongoing DARPA project Low Resource Languages for Emergent Incidents (LORELEI) aims to further language processing technologies for low resource languages in the context of such a humanitarian crisis. In this work, we describe our submission for the 2019 Sentiment, Emotion and Cognitive state (SEC) pilot task of the LORELEI project. We describe a collection of sentiment analysis systems included in our submission along with the features extracted. Our fielded systems obtained the best results in both English and Spanish language evaluations of the SEC pilot task.
Tasks	Sentiment Analysis
Published	2019-05-01
URL	http://arxiv.org/abs/1905.00472v1
PDF	http://arxiv.org/pdf/1905.00472v1.pdf
PWC	https://paperswithcode.com/paper/a-system-for-the-2019-sentiment-emotion-and
Repo
Framework

Supervised Uncertainty Quantification for Segmentation with Multiple Annotations


Title	Supervised Uncertainty Quantification for Segmentation with Multiple Annotations
Authors	Shi Hu, Daniel Worrall, Stefan Knegt, Bas Veeling, Henkjan Huisman, Max Welling
Abstract	The accurate estimation of predictive uncertainty carries importance in medical scenarios such as lung node segmentation. Unfortunately, most existing works on predictive uncertainty do not return calibrated uncertainty estimates, which could be used in practice. In this work we exploit multi-grader annotation variability as a source of ‘groundtruth’ aleatoric uncertainty, which can be treated as a target in a supervised learning problem. We combine this groundtruth uncertainty with a Probabilistic U-Net and test on the LIDC-IDRI lung nodule CT dataset and MICCAI2012 prostate MRI dataset. We find that we are able to improve predictive uncertainty estimates. We also find that we can improve sample accuracy and sample diversity.
Tasks
Published	2019-07-03
URL	https://arxiv.org/abs/1907.01949v1
PDF	https://arxiv.org/pdf/1907.01949v1.pdf
PWC	https://paperswithcode.com/paper/supervised-uncertainty-quantification-for
Repo
Framework

NONOTO: A Model-agnostic Web Interface for Interactive Music Composition by Inpainting


Title	NONOTO: A Model-agnostic Web Interface for Interactive Music Composition by Inpainting
Authors	Théis Bazin, Gaëtan Hadjeres
Abstract	Inpainting-based generative modeling allows for stimulating human-machine interactions by letting users perform stylistically coherent local editions to an object using a statistical model. We present NONOTO, a new interface for interactive music generation based on inpainting models. It is aimed both at researchers, by offering a simple and flexible API allowing them to connect their own models with the interface, and at musicians by providing industry-standard features such as audio playback, real-time MIDI output and straightforward synchronization with DAWs using Ableton Link.
Tasks	Music Generation
Published	2019-07-23
URL	https://arxiv.org/abs/1907.10380v1
PDF	https://arxiv.org/pdf/1907.10380v1.pdf
PWC	https://paperswithcode.com/paper/nonoto-a-model-agnostic-web-interface-for
Repo
Framework

Assessment of gait normality using a depth camera and mirrors


Title	Assessment of gait normality using a depth camera and mirrors
Authors	Trong Nguyen Nguyen, Huu Hung Huynh, Jean Meunier
Abstract	This paper presents an initial work on assessment of gait normality in which the human body motion is represented by a sequence of enhanced depth maps. The input data is provided by a system consisting of a Time-of-Flight (ToF) depth camera and two mirrors. This approach proposes two feature types to describe characteristics of localized points of interest and the level of posture symmetry. These two features are processed on a sequence of enhanced depth maps with the support of a sliding window to provide two corresponding scores. The gait assessment is finally performed based on a weighted combination of these two scores. The evaluation is performed by experimenting on 6 simulated abnormal gaits.
Tasks
Published	2019-08-17
URL	https://arxiv.org/abs/1908.07418v1
PDF	https://arxiv.org/pdf/1908.07418v1.pdf
PWC	https://paperswithcode.com/paper/assessment-of-gait-normality-using-a-depth
Repo
Framework

CAQL: Continuous Action Q-Learning


Title	CAQL: Continuous Action Q-Learning
Authors	Moonkyung Ryu, Yinlam Chow, Ross Anderson, Christian Tjandraatmadja, Craig Boutilier
Abstract	Value-based reinforcement learning (RL) methods like Q-learning have shown success in a variety of domains. One challenge in applying Q-learning to continuous-action RL problems, however, is the continuous action maximization (max-Q) required for optimal Bellman backup. In this work, we develop CAQL, a (class of) algorithm(s) for continuous-action Q-learning that can use several plug-and-play optimizers for the max-Q problem. Leveraging recent optimization results for deep neural networks, we show that max-Q can be solved optimally using mixed-integer programming (MIP). When the Q-function representation has sufficient power, MIP-based optimization gives rise to better policies and is more robust than approximate methods (e.g., gradient ascent, cross-entropy search). We further develop several techniques to accelerate inference in CAQL, which despite their approximate nature, perform well. We compare CAQL with state-of-the-art RL algorithms on benchmark continuous-control problems that have different degrees of action constraints and show that CAQL outperforms policy-based methods in heavily constrained environments, often dramatically.
Tasks	Continuous Control, Q-Learning
Published	2019-09-26
URL	https://arxiv.org/abs/1909.12397v3
PDF	https://arxiv.org/pdf/1909.12397v3.pdf
PWC	https://paperswithcode.com/paper/caql-continuous-action-q-learning
Repo
Framework

Do Design Metrics Capture Developers Perception of Quality? An Empirical Study on Self-Affirmed Refactoring Activities


Title	Do Design Metrics Capture Developers Perception of Quality? An Empirical Study on Self-Affirmed Refactoring Activities
Authors	Eman Abdullah AlOmar, Mohamed Wiem Mkaouer, Ali Ouni, Marouane Kessentini
Abstract	Background. Refactoring is a critical task in software maintenance and is generally performed to enforce the best design and implementation practices or to cope with design defects. Several studies attempted to detect refactoring activities through mining software repositories allowing to collect, analyze and get actionable data-driven insights about refactoring practices within software projects. Aim. We aim at identifying, among the various quality models presented in the literature, the ones that are more in-line with the developer’s vision of quality optimization, when they explicitly mention that they are refactoring to improve them. Method. We extract a large corpus of design-related refactoring activities that are applied and documented by developers during their daily changes from 3,795 curated open source Java projects. In particular, we extract a large-scale corpus of structural metrics and anti-pattern enhancement changes, from which we identify 1,245 quality improvement commits with their corresponding refactoring operations, as perceived by software engineers. Thereafter, we empirically analyze the impact of these refactoring operations on a set of common state-of-the-art design quality metrics. Results. The statistical analysis of the obtained results shows that (i) a few state-of-the-art metrics are more popular than others; and (ii) some metrics are being more emphasized than others. Conclusions. We verify that there are a variety of structural metrics that can represent the internal quality attributes with different degrees of improvement and degradation of software quality. Most of the metrics that are mapped to the main quality attributes do capture developer intentions of quality improvement reported in the commit messages.
Tasks
Published	2019-07-10
URL	https://arxiv.org/abs/1907.04797v1
PDF	https://arxiv.org/pdf/1907.04797v1.pdf
PWC	https://paperswithcode.com/paper/do-design-metrics-capture-developers
Repo
Framework