July 28, 2019

3130 words 15 mins read

Paper Group ANR 175

Data Augmentation for Low-Resource Neural Machine Translation. The Effect of Communication on Noncooperative Multiplayer Multi-Armed Bandit Problems. Line Profile Based Segmentation Algorithm for Touching Corn Kernels. Inverse Reinforcement Learning in Large State Spaces via Function Approximation. Strategies for Conceptual Change in Convolutional …

Data Augmentation for Low-Resource Neural Machine Translation


Title	Data Augmentation for Low-Resource Neural Machine Translation
Authors	Marzieh Fadaee, Arianna Bisazza, Christof Monz
Abstract	The quality of a Neural Machine Translation system depends substantially on the availability of sizable parallel corpora. For low-resource language pairs this is not the case, resulting in poor translation quality. Inspired by work in computer vision, we propose a novel data augmentation approach that targets low-frequency words by generating new sentence pairs containing rare words in new, synthetically created contexts. Experimental results on simulated low-resource settings show that our method improves translation quality by up to 2.9 BLEU points over the baseline and up to 3.2 BLEU over back-translation.
Tasks	Data Augmentation, Low-Resource Neural Machine Translation, Machine Translation
Published	2017-05-01
URL	http://arxiv.org/abs/1705.00440v1
PDF	http://arxiv.org/pdf/1705.00440v1.pdf
PWC	https://paperswithcode.com/paper/data-augmentation-for-low-resource-neural
Repo
Framework

The Effect of Communication on Noncooperative Multiplayer Multi-Armed Bandit Problems


Title	The Effect of Communication on Noncooperative Multiplayer Multi-Armed Bandit Problems
Authors	Noyan Evirgen, Alper Kose
Abstract	We consider decentralized stochastic multi-armed bandit problem with multiple players in the case of different communication probabilities between players. Each player makes a decision of pulling an arm without cooperation while aiming to maximize his or her reward but informs his or her neighbors in the end of every turn about the arm he or she pulled and the reward he or she got. Neighbors of players are determined according to an Erdos-Renyi graph with which is reproduced in the beginning of every turn. We consider i.i.d. rewards generated by a Bernoulli distribution and assume that players are unaware about the arms’ probability distributions and their mean values. In case of a collision, we assume that only one of the players who is randomly chosen gets the reward where the others get zero reward. We study the effects of connectivity, the degree of communication between players, on the cumulative regret using well-known algorithms UCB1, epsilon-Greedy and Thompson Sampling.
Tasks
Published	2017-11-05
URL	http://arxiv.org/abs/1711.01628v1
PDF	http://arxiv.org/pdf/1711.01628v1.pdf
PWC	https://paperswithcode.com/paper/the-effect-of-communication-on-noncooperative
Repo
Framework

Line Profile Based Segmentation Algorithm for Touching Corn Kernels


Title	Line Profile Based Segmentation Algorithm for Touching Corn Kernels
Authors	Ali Mahdi, Jun Qin
Abstract	Image segmentation of touching objects plays a key role in providing accurate classification for computer vision technologies. A new line profile based imaging segmentation algorithm has been developed to provide a robust and accurate segmentation of a group of touching corns. The performance of the line profile based algorithm has been compared to a watershed based imaging segmentation algorithm. Both algorithms are tested on three different patterns of images, which are isolated corns, single-lines, and random distributed formations. The experimental results show that the algorithm can segment a large number of touching corn kernels efficiently and accurately.
Tasks	Semantic Segmentation
Published	2017-06-01
URL	http://arxiv.org/abs/1706.00396v3
PDF	http://arxiv.org/pdf/1706.00396v3.pdf
PWC	https://paperswithcode.com/paper/line-profile-based-segmentation-algorithm-for
Repo
Framework

Inverse Reinforcement Learning in Large State Spaces via Function Approximation


Title	Inverse Reinforcement Learning in Large State Spaces via Function Approximation
Authors	Kun Li, Joel W. Burdick
Abstract	This paper introduces a new method for inverse reinforcement learning in large-scale and high-dimensional state spaces. To avoid solving the computationally expensive reinforcement learning problems in reward learning, we propose a function approximation method to ensure that the Bellman Optimality Equation always holds, and then estimate a function to maximize the likelihood of the observed motion. The time complexity of the proposed method is linearly proportional to the cardinality of the action set, thus it can handle large state spaces efficiently. We test the proposed method in a simulated environment, and show that it is more accurate than existing methods and significantly better in scalability. We also show that the proposed method can extend many existing methods to high-dimensional state spaces. We then apply the method to evaluating the effect of rehabilitative stimulations on patients with spinal cord injuries based on the observed patient motions.
Tasks
Published	2017-07-28
URL	http://arxiv.org/abs/1707.09394v3
PDF	http://arxiv.org/pdf/1707.09394v3.pdf
PWC	https://paperswithcode.com/paper/inverse-reinforcement-learning-in-large-state
Repo
Framework

Strategies for Conceptual Change in Convolutional Neural Networks


Title	Strategies for Conceptual Change in Convolutional Neural Networks
Authors	Maarten Grachten, Carlos Eduardo Cancino Chacón
Abstract	A remarkable feature of human beings is their capacity for creative behaviour, referring to their ability to react to problems in ways that are novel, surprising, and useful. Transformational creativity is a form of creativity where the creative behaviour is induced by a transformation of the actor’s conceptual space, that is, the representational system with which the actor interprets its environment. In this report, we focus on ways of adapting systems of learned representations as they switch from performing one task to performing another. We describe an experimental comparison of multiple strategies for adaptation of learned features, and evaluate how effectively each of these strategies realizes the adaptation, in terms of the amount of training, and in terms of their ability to cope with restricted availability of training data. We show, among other things, that across handwritten digits, natural images, and classical music, adaptive strategies are systematically more effective than a baseline method that starts learning from scratch.
Tasks
Published	2017-11-05
URL	https://arxiv.org/abs/1711.01634v2
PDF	https://arxiv.org/pdf/1711.01634v2.pdf
PWC	https://paperswithcode.com/paper/strategies-for-conceptual-change-in
Repo
Framework

Design of PI Controller for Automatic Generation Control of Multi Area Interconnected Power System using Bacterial Foraging Optimization


Title	Design of PI Controller for Automatic Generation Control of Multi Area Interconnected Power System using Bacterial Foraging Optimization
Authors	Naresh Kumari, Nitin Malik, A. N. Jha, Gaddam Mallesham
Abstract	The system comprises of three interconnected power system networks based on thermal, wind and hydro power generation. The load variation in any one of the network results in frequency deviation in all the connected systems.The PI controllers have been connected separately with each system for the frequency control and the gains (Kp and Ki) of all the controllers have been optimized along with frequency bias (Bi) and speed regulation parameter (Ri). The computationally intelligent techniques like bacterial foraging optimization (BFO) and particle swarm optimization (PSO) have been applied for the tuning of controller gains along with variable parameters Bi and Ri. The gradient descent (GD) based conventional method has also been applied for optimizing the parameters Kp, Ki,Bi and Ri.The frequency responses are obtained with all the methods. The performance index chosen is the integral square error (ISE). The settling time, peak overshoot and peak undershoot of all the frequency responses on applying three optimization techniques have been compared. It has been observed that the peak overshoot and peak undershoot significantly reduce with BFO technique followed by the PSO and GD techniques. While obtaining such optimum response the settling time is increased marginally with bacterial foraging technique due to large number of mathematical equations used for the computation in BFO. The comparison of frequency response using three techniques show the superiority of BFO over the PSO and GD techniques. The designing of the system and tuning of the parameters with three techniques has been done in MATLAB/SIMULINK environment.
Tasks
Published	2017-01-26
URL	http://arxiv.org/abs/1701.08081v1
PDF	http://arxiv.org/pdf/1701.08081v1.pdf
PWC	https://paperswithcode.com/paper/design-of-pi-controller-for-automatic
Repo
Framework

Correlational Dueling Bandits with Application to Clinical Treatment in Large Decision Spaces


Title	Correlational Dueling Bandits with Application to Clinical Treatment in Large Decision Spaces
Authors	Yanan Sui, Yisong Yue, Joel W. Burdick
Abstract	We consider sequential decision making under uncertainty, where the goal is to optimize over a large decision space using noisy comparative feedback. This problem can be formulated as a $K$-armed Dueling Bandits problem where $K$ is the total number of decisions. When $K$ is very large, existing dueling bandits algorithms suffer huge cumulative regret before converging on the optimal arm. This paper studies the dueling bandits problem with a large number of arms that exhibit a low-dimensional correlation structure. Our problem is motivated by a clinical decision making process in large decision space. We propose an efficient algorithm CorrDuel which optimizes the exploration/exploitation tradeoff in this large decision space of clinical treatments. More broadly, our approach can be applied to other sequential decision problems with large and structured decision spaces. We derive regret bounds, and evaluate performance in simulation experiments as well as on a live clinical trial of therapeutic spinal cord stimulation. To our knowledge, this marks the first time an online learning algorithm was applied towards spinal cord injury treatments. Our experimental results show the effectiveness and efficiency of our approach.
Tasks	Decision Making, Decision Making Under Uncertainty
Published	2017-07-08
URL	http://arxiv.org/abs/1707.02375v1
PDF	http://arxiv.org/pdf/1707.02375v1.pdf
PWC	https://paperswithcode.com/paper/correlational-dueling-bandits-with
Repo
Framework

Multi-Task Video Captioning with Video and Entailment Generation


Title	Multi-Task Video Captioning with Video and Entailment Generation
Authors	Ramakanth Pasunuru, Mohit Bansal
Abstract	Video captioning, the task of describing the content of a video, has seen some promising improvements in recent years with sequence-to-sequence models, but accurately learning the temporal and logical dynamics involved in the task still remains a challenge, especially given the lack of sufficient annotated data. We improve video captioning by sharing knowledge with two related directed-generation tasks: a temporally-directed unsupervised video prediction task to learn richer context-aware video encoder representations, and a logically-directed language entailment generation task to learn better video-entailed caption decoder representations. For this, we present a many-to-many multi-task learning model that shares parameters across the encoders and decoders of the three tasks. We achieve significant improvements and the new state-of-the-art on several standard video captioning datasets using diverse automatic and human evaluations. We also show mutual multi-task improvements on the entailment generation task.
Tasks	Multi-Task Learning, Video Captioning, Video Prediction
Published	2017-04-24
URL	http://arxiv.org/abs/1704.07489v2
PDF	http://arxiv.org/pdf/1704.07489v2.pdf
PWC	https://paperswithcode.com/paper/multi-task-video-captioning-with-video-and
Repo
Framework

Calendar.help: Designing a Workflow-Based Scheduling Agent with Humans in the Loop


Title	Calendar.help: Designing a Workflow-Based Scheduling Agent with Humans in the Loop
Authors	Justin Cranshaw, Emad Elwany, Todd Newman, Rafal Kocielnik, Bowen Yu, Sandeep Soni, Jaime Teevan, Andrés Monroy-Hernández
Abstract	Although information workers may complain about meetings, they are an essential part of their work life. Consequently, busy people spend a significant amount of time scheduling meetings. We present Calendar.help, a system that provides fast, efficient scheduling through structured workflows. Users interact with the system via email, delegating their scheduling needs to the system as if it were a human personal assistant. Common scheduling scenarios are broken down using well-defined workflows and completed as a series of microtasks that are automated when possible and executed by a human otherwise. Unusual scenarios fall back to a trained human assistant who executes them as unstructured macrotasks. We describe the iterative approach we used to develop Calendar.help, and share the lessons learned from scheduling thousands of meetings during a year of real-world deployments. Our findings provide insight into how complex information tasks can be broken down into repeatable components that can be executed efficiently to improve productivity.
Tasks
Published	2017-03-24
URL	http://arxiv.org/abs/1703.08428v1
PDF	http://arxiv.org/pdf/1703.08428v1.pdf
PWC	https://paperswithcode.com/paper/calendarhelp-designing-a-workflow-based
Repo
Framework

Binary Patterns Encoded Convolutional Neural Networks for Texture Recognition and Remote Sensing Scene Classification


Title	Binary Patterns Encoded Convolutional Neural Networks for Texture Recognition and Remote Sensing Scene Classification
Authors	Rao Muhammad Anwer, Fahad Shahbaz Khan, Joost van de Weijer, Matthieu Molinier, Jorma Laaksonen
Abstract	Designing discriminative powerful texture features robust to realistic imaging conditions is a challenging computer vision problem with many applications, including material recognition and analysis of satellite or aerial imagery. In the past, most texture description approaches were based on dense orderless statistical distribution of local features. However, most recent approaches to texture recognition and remote sensing scene classification are based on Convolutional Neural Networks (CNNs). The d facto practice when learning these CNN models is to use RGB patches as input with training performed on large amounts of labeled data (ImageNet). In this paper, we show that Binary Patterns encoded CNN models, codenamed TEX-Nets, trained using mapped coded images with explicit texture information provide complementary information to the standard RGB deep models. Additionally, two deep architectures, namely early and late fusion, are investigated to combine the texture and color information. To the best of our knowledge, we are the first to investigate Binary Patterns encoded CNNs and different deep network fusion architectures for texture recognition and remote sensing scene classification. We perform comprehensive experiments on four texture recognition datasets and four remote sensing scene classification benchmarks: UC-Merced with 21 scene categories, WHU-RS19 with 19 scene classes, RSSCN7 with 7 categories and the recently introduced large scale aerial image dataset (AID) with 30 aerial scene types. We demonstrate that TEX-Nets provide complementary information to standard RGB deep model of the same network architecture. Our late fusion TEX-Net architecture always improves the overall performance compared to the standard RGB network on both recognition problems. Our final combination outperforms the state-of-the-art without employing fine-tuning or ensemble of RGB network architectures.
Tasks	Material Recognition, Scene Classification
Published	2017-06-05
URL	http://arxiv.org/abs/1706.01171v2
PDF	http://arxiv.org/pdf/1706.01171v2.pdf
PWC	https://paperswithcode.com/paper/binary-patterns-encoded-convolutional-neural
Repo
Framework

Retrosynthetic reaction prediction using neural sequence-to-sequence models


Title	Retrosynthetic reaction prediction using neural sequence-to-sequence models
Authors	Bowen Liu, Bharath Ramsundar, Prasad Kawthekar, Jade Shi, Joseph Gomes, Quang Luu Nguyen, Stephen Ho, Jack Sloane, Paul Wender, Vijay Pande
Abstract	We describe a fully data driven model that learns to perform a retrosynthetic reaction prediction task, which is treated as a sequence-to-sequence mapping problem. The end-to-end trained model has an encoder-decoder architecture that consists of two recurrent neural networks, which has previously shown great success in solving other sequence-to-sequence prediction tasks such as machine translation. The model is trained on 50,000 experimental reaction examples from the United States patent literature, which span 10 broad reaction types that are commonly used by medicinal chemists. We find that our model performs comparably with a rule-based expert system baseline model, and also overcomes certain limitations associated with rule-based expert systems and with any machine learning approach that contains a rule-based expert system component. Our model provides an important first step towards solving the challenging problem of computational retrosynthetic analysis.
Tasks	Machine Translation
Published	2017-06-06
URL	http://arxiv.org/abs/1706.01643v1
PDF	http://arxiv.org/pdf/1706.01643v1.pdf
PWC	https://paperswithcode.com/paper/retrosynthetic-reaction-prediction-using
Repo
Framework

Riemannian-geometry-based modeling and clustering of network-wide non-stationary time series: The brain-network case


Title	Riemannian-geometry-based modeling and clustering of network-wide non-stationary time series: The brain-network case
Authors	Konstantinos Slavakis, Shiva Salsabilian, David S. Wack, Sarah F. Muldoon, Henry E. Baidoo-Williams, Jean M. Vettel, Matthew Cieslak, Scott T. Grafton
Abstract	This paper advocates Riemannian multi-manifold modeling in the context of network-wide non-stationary time-series analysis. Time-series data, collected sequentially over time and across a network, yield features which are viewed as points in or close to a union of multiple submanifolds of a Riemannian manifold, and distinguishing disparate time series amounts to clustering multiple Riemannian submanifolds. To support the claim that exploiting the latent Riemannian geometry behind many statistical features of time series is beneficial to learning from network data, this paper focuses on brain networks and puts forth two feature-generation schemes for network-wide dynamic time series. The first is motivated by Granger-causality arguments and uses an auto-regressive moving average model to map low-rank linear vector subspaces, spanned by column vectors of appropriately defined observability matrices, to points into the Grassmann manifold. The second utilizes (non-linear) dependencies among network nodes by introducing kernel-based partial correlations to generate points in the manifold of positive-definite matrices. Capitilizing on recently developed research on clustering Riemannian submanifolds, an algorithm is provided for distinguishing time series based on their geometrical properties, revealed within Riemannian feature spaces. Extensive numerical tests demonstrate that the proposed framework outperforms classical and state-of-the-art techniques in clustering brain-network states/structures hidden beneath synthetic fMRI time series and brain-activity signals generated from real brain-network structural connectivity matrices.
Tasks	Time Series, Time Series Analysis
Published	2017-01-26
URL	http://arxiv.org/abs/1701.07767v1
PDF	http://arxiv.org/pdf/1701.07767v1.pdf
PWC	https://paperswithcode.com/paper/riemannian-geometry-based-modeling-and
Repo
Framework

Face R-CNN


Title	Face R-CNN
Authors	Hao Wang, Zhifeng Li, Xing Ji, Yitong Wang
Abstract	Faster R-CNN is one of the most representative and successful methods for object detection, and has been becoming increasingly popular in various objection detection applications. In this report, we propose a robust deep face detection approach based on Faster R-CNN. In our approach, we exploit several new techniques including new multi-task loss function design, online hard example mining, and multi-scale training strategy to improve Faster R-CNN in multiple aspects. The proposed approach is well suited for face detection, so we call it Face R-CNN. Extensive experiments are conducted on two most popular and challenging face detection benchmarks, FDDB and WIDER FACE, to demonstrate the superiority of the proposed approach over state-of-the-arts.
Tasks	Face Detection, Object Detection
Published	2017-06-04
URL	http://arxiv.org/abs/1706.01061v1
PDF	http://arxiv.org/pdf/1706.01061v1.pdf
PWC	https://paperswithcode.com/paper/face-r-cnn
Repo
Framework

Phase Retrieval via Randomized Kaczmarz: Theoretical Guarantees


Title	Phase Retrieval via Randomized Kaczmarz: Theoretical Guarantees
Authors	Yan Shuo Tan, Roman Vershynin
Abstract	We consider the problem of phase retrieval, i.e. that of solving systems of quadratic equations. A simple variant of the randomized Kaczmarz method was recently proposed for phase retrieval, and it was shown numerically to have a computational edge over state-of-the-art Wirtinger flow methods. In this paper, we provide the first theoretical guarantee for the convergence of the randomized Kaczmarz method for phase retrieval. We show that it is sufficient to have as many Gaussian measurements as the dimension, up to a constant factor. Along the way, we introduce a sufficient condition on measurement sets for which the randomized Kaczmarz method is guaranteed to work. We show that Gaussian sampling vectors satisfy this property with high probability; this is proved using a chaining argument coupled with bounds on VC dimension and metric entropy.
Tasks
Published	2017-06-30
URL	http://arxiv.org/abs/1706.09993v2
PDF	http://arxiv.org/pdf/1706.09993v2.pdf
PWC	https://paperswithcode.com/paper/phase-retrieval-via-randomized-kaczmarz
Repo
Framework

Iterative Machine Learning for Precision Trajectory Tracking with Series Elastic Actuators


Title	Iterative Machine Learning for Precision Trajectory Tracking with Series Elastic Actuators
Authors	Nathan Banka, W. Tony Piaskowy, Joseph Garbini, Santosh Devasia
Abstract	When robots operate in unknown environments small errors in postions can lead to large variations in the contact forces, especially with typical high-impedance designs. This can potentially damage the surroundings and/or the robot. Series elastic actuators (SEAs) are a popular way to reduce the output impedance of a robotic arm to improve control authority over the force exerted on the environment. However this increased control over forces with lower impedance comes at the cost of lower positioning precision and bandwidth. This article examines the use of an iteratively-learned feedforward command to improve position tracking when using SEAs. Over each iteration, the output responses of the system to the quantized inputs are used to estimate a linearized local system models. These estimated models are obtained using a complex-valued Gaussian Process Regression (cGPR) technique and then, used to generate a new feedforward input command based on the previous iteration’s error. This article illustrates this iterative machine learning (IML) technique for a two degree of freedom (2-DOF) robotic arm, and demonstrates successful convergence of the IML approach to reduce the tracking error.
Tasks
Published	2017-10-05
URL	http://arxiv.org/abs/1710.09691v1
PDF	http://arxiv.org/pdf/1710.09691v1.pdf
PWC	https://paperswithcode.com/paper/iterative-machine-learning-for-precision
Repo
Framework