February 1, 2020

3084 words 15 mins read

Paper Group AWR 147

Tactical Rewind: Self-Correction via Backtracking in Vision-and-Language Navigation. Cascaded Partial Decoder for Fast and Accurate Salient Object Detection. Predicting Rare Events in Multiscale Dynamical Systems using Machine Learning. Data-efficient Learning of Morphology and Controller for a Microrobot. Joint Extraction of Entities and Relations …


Title	Tactical Rewind: Self-Correction via Backtracking in Vision-and-Language Navigation
Authors	Liyiming Ke, Xiujun Li, Yonatan Bisk, Ari Holtzman, Zhe Gan, Jingjing Liu, Jianfeng Gao, Yejin Choi, Siddhartha Srinivasa
Abstract	We present the Frontier Aware Search with backTracking (FAST) Navigator, a general framework for action decoding, that achieves state-of-the-art results on the Room-to-Room (R2R) Vision-and-Language navigation challenge of Anderson et. al. (2018). Given a natural language instruction and photo-realistic image views of a previously unseen environment, the agent was tasked with navigating from source to target location as quickly as possible. While all current approaches make local action decisions or score entire trajectories using beam search, ours balances local and global signals when exploring an unobserved environment. Importantly, this lets us act greedily but use global signals to backtrack when necessary. Applying FAST framework to existing state-of-the-art models achieved a 17% relative gain, an absolute 6% gain on Success rate weighted by Path Length (SPL).
Tasks	Vision-Language Navigation
Published	2019-03-06
URL	http://arxiv.org/abs/1903.02547v2
PDF	http://arxiv.org/pdf/1903.02547v2.pdf
PWC	https://paperswithcode.com/paper/tactical-rewind-self-correction-via
Repo	https://github.com/Kelym/FAST
Framework	pytorch

Cascaded Partial Decoder for Fast and Accurate Salient Object Detection


Title	Cascaded Partial Decoder for Fast and Accurate Salient Object Detection
Authors	Zhe Wu, Li Su, Qingming Huang
Abstract	Existing state-of-the-art salient object detection networks rely on aggregating multi-level features of pre-trained convolutional neural networks (CNNs). Compared to high-level features, low-level features contribute less to performance but cost more computations because of their larger spatial resolutions. In this paper, we propose a novel Cascaded Partial Decoder (CPD) framework for fast and accurate salient object detection. On the one hand, the framework constructs partial decoder which discards larger resolution features of shallower layers for acceleration. On the other hand, we observe that integrating features of deeper layers obtain relatively precise saliency map. Therefore we directly utilize generated saliency map to refine the features of backbone network. This strategy efficiently suppresses distractors in the features and significantly improves their representation ability. Experiments conducted on five benchmark datasets exhibit that the proposed model not only achieves state-of-the-art performance but also runs much faster than existing models. Besides, the proposed framework is further applied to improve existing multi-level feature aggregation models and significantly improve their efficiency and accuracy.
Tasks	Object Detection, Salient Object Detection
Published	2019-04-18
URL	http://arxiv.org/abs/1904.08739v1
PDF	http://arxiv.org/pdf/1904.08739v1.pdf
PWC	https://paperswithcode.com/paper/cascaded-partial-decoder-for-fast-and
Repo	https://github.com/wuzhe71/CPD
Framework	pytorch

Predicting Rare Events in Multiscale Dynamical Systems using Machine Learning


Title	Predicting Rare Events in Multiscale Dynamical Systems using Machine Learning
Authors	Soon Hoe Lim, Ludovico Theo Giorgini, Woosok Moon, J. S. Wettlaufer
Abstract	We study the problem of rare event prediction for a class of slow-fast nonlinear dynamical systems. The state of the system of interest is described by a slow process, whereas a faster process drives its evolution. By taking advantage of recent advances in machine learning, we present a data-driven method to predict the future evolution of the state. We show that our method is capable of predicting a rare event at least several time steps in advance. We demonstrate our method using numerical experiments on three examples of systems, ranging from low dimensional to high dimensional. We discuss the mathematical and broader implications of our results.
Tasks
Published	2019-08-10
URL	https://arxiv.org/abs/1908.03771v2
PDF	https://arxiv.org/pdf/1908.03771v2.pdf
PWC	https://paperswithcode.com/paper/predicting-rare-events-in-multiscale
Repo	https://github.com/shoelim/predicting_rare_events_multiscale_systems
Framework	none

Data-efficient Learning of Morphology and Controller for a Microrobot


Title	Data-efficient Learning of Morphology and Controller for a Microrobot
Authors	Thomas Liao, Grant Wang, Brian Yang, Rene Lee, Kristofer Pister, Sergey Levine, Roberto Calandra
Abstract	Robot design is often a slow and difficult process requiring the iterative construction and testing of prototypes, with the goal of sequentially optimizing the design. For most robots, this process is further complicated by the need, when validating the capabilities of the hardware to solve the desired task, to already have an appropriate controller, which is in turn designed and tuned for the specific hardware. In this paper, we propose a novel approach, HPC-BBO, to efficiently and automatically design hardware configurations, and evaluate them by also automatically tuning the corresponding controller. HPC-BBO is based on a hierarchical Bayesian optimization process which iteratively optimizes morphology configurations (based on the performance of the previous designs during the controller learning process) and subsequently learns the corresponding controllers (exploiting the knowledge collected from optimizing for previous morphologies). Moreover, HPC-BBO can select a “batch” of multiple morphology designs at once, thus parallelizing hardware validation and reducing the number of time-consuming production cycles. We validate HPC-BBO on the design of the morphology and controller for a simulated 6-legged microrobot. Experimental results show that HPC-BBO outperforms multiple competitive baselines, and yields a $360%$ reduction in production cycles over standard Bayesian optimization, thus reducing the hypothetical manufacturing time of our microrobot from 21 to 4 months.
Tasks
Published	2019-05-03
URL	https://arxiv.org/abs/1905.01334v1
PDF	https://arxiv.org/pdf/1905.01334v1.pdf
PWC	https://paperswithcode.com/paper/data-efficient-learning-of-morphology-and
Repo	https://github.com/tholiao/learning-morph-and-ctrl
Framework	none

Joint Extraction of Entities and Relations Based on a Novel Decomposition Strategy


Title	Joint Extraction of Entities and Relations Based on a Novel Decomposition Strategy
Authors	Bowen Yu, Zhenyu Zhang, Xiaobo Shu, Yubin Wang, Tingwen Liu, Bin Wang, Sujian Li
Abstract	Joint extraction of entities and relations aims to detect entity pairs along with their relations using a single model. Prior work typically solves this task in the extract-then-classify or unified labeling manner. However, these methods either suffer from the redundant entity pairs, or ignore the important inner structure in the process of extracting entities and relations. To address these limitations, in this paper, we first decompose the joint extraction task into two interrelated subtasks, namely HE extraction and TER extraction. The former subtask is to distinguish all head-entities that may be involved with target relations, and the latter is to identify corresponding tail-entities and relations for each extracted head-entity. Next, these two subtasks are further deconstructed into several sequence labeling problems based on our proposed span-based tagging scheme, which are conveniently solved by a hierarchical boundary tagger and a multi-span decoding algorithm. Owing to the reasonable decomposition strategy, our model can fully capture the semantic interdependency between different steps, as well as reduce noise from irrelevant entity pairs. Experimental results show that our method outperforms previous work by 5.2%, 5.9% and 21.5% (F1 score), achieving a new state-of-the-art on three public datasets
Tasks	Relation Extraction
Published	2019-09-10
URL	https://arxiv.org/abs/1909.04273v3
PDF	https://arxiv.org/pdf/1909.04273v3.pdf
PWC	https://paperswithcode.com/paper/joint-extraction-of-entities-and-relations-3
Repo	https://github.com/yubowen-ph/JointER
Framework	pytorch

Symbolic Regression for Constructing Analytic Models in Reinforcement Learning


Title	Symbolic Regression for Constructing Analytic Models in Reinforcement Learning
Authors	Erik Derner, Jiří Kubalík, Nicola Ancona, Robert Babuška
Abstract	Reinforcement learning (RL) is a widely used approach for controlling systems with unknown or time-varying dynamics. Even though RL does not require a model of the system, it is known to be faster and safer when using models learned online. We propose to employ symbolic regression (SR) to construct parsimonious process models described by analytic equations for real-time RL control. We have tested our method with two different state-of-the-art SR algorithms which automatically search for equations that fit the measured data. In addition to the standard problem formulation in the state-space domain, we show how the method can also be applied to input-output models of the NARX (nonlinear autoregressive with exogenous input) type. We present the approach on three simulated examples with up to 14-dimensional state space: an inverted pendulum, a mobile robot, and a biped walking robot. A comparison with deep neural networks and local linear regression shows that SR in most cases outperforms these commonly used alternative methods. We demonstrate on a real pendulum system that the analytic model found enables RL to successfully perform the swing-up task, based on a model constructed from only 100 data samples.
Tasks
Published	2019-03-27
URL	http://arxiv.org/abs/1903.11483v1
PDF	http://arxiv.org/pdf/1903.11483v1.pdf
PWC	https://paperswithcode.com/paper/symbolic-regression-for-constructing-analytic
Repo	https://github.com/erik-derner/symbolic-regression
Framework	none

Racial Bias in Hate Speech and Abusive Language Detection Datasets


Title	Racial Bias in Hate Speech and Abusive Language Detection Datasets
Authors	Thomas Davidson, Debasmita Bhattacharya, Ingmar Weber
Abstract	Technologies for abusive language detection are being developed and applied with little consideration of their potential biases. We examine racial bias in five different sets of Twitter data annotated for hate speech and abusive language. We train classifiers on these datasets and compare the predictions of these classifiers on tweets written in African-American English with those written in Standard American English. The results show evidence of systematic racial bias in all datasets, as classifiers trained on them tend to predict that tweets written in African-American English are abusive at substantially higher rates. If these abusive language detection systems are used in the field they will therefore have a disproportionate negative impact on African-American social media users. Consequently, these systems may discriminate against the groups who are often the targets of the abuse we are trying to detect.
Tasks
Published	2019-05-29
URL	https://arxiv.org/abs/1905.12516v1
PDF	https://arxiv.org/pdf/1905.12516v1.pdf
PWC	https://paperswithcode.com/paper/racial-bias-in-hate-speech-and-abusive
Repo	https://github.com/t-davidson/hate-speech-and-offensive-language
Framework	none

Zero-shot Knowledge Transfer via Adversarial Belief Matching


Title	Zero-shot Knowledge Transfer via Adversarial Belief Matching
Authors	Paul Micaelli, Amos Storkey
Abstract	Performing knowledge transfer from a large teacher network to a smaller student is a popular task in modern deep learning applications. However, due to growing dataset sizes and stricter privacy regulations, it is increasingly common not to have access to the data that was used to train the teacher. We propose a novel method which trains a student to match the predictions of its teacher without using any data or metadata. We achieve this by training an adversarial generator to search for images on which the student poorly matches the teacher, and then using them to train the student. Our resulting student closely approximates its teacher for simple datasets like SVHN, and on CIFAR10 we improve on the state-of-the-art for few-shot distillation (with 100 images per class), despite using no data. Finally, we also propose a metric to quantify the degree of belief matching between teacher and student in the vicinity of decision boundaries, and observe a significantly higher match between our zero-shot student and the teacher, than between a student distilled with real data and the teacher. Code available at: https://github.com/polo5/ZeroShotKnowledgeTransfer
Tasks	Transfer Learning
Published	2019-05-23
URL	https://arxiv.org/abs/1905.09768v4
PDF	https://arxiv.org/pdf/1905.09768v4.pdf
PWC	https://paperswithcode.com/paper/zero-shot-knowledge-transfer-via-adversarial
Repo	https://github.com/VainF/DFAD
Framework	pytorch

Semi-Supervised Learning with Scarce Annotations


Title	Semi-Supervised Learning with Scarce Annotations
Authors	Sylvestre-Alvise Rebuffi, Sebastien Ehrhardt, Kai Han, Andrea Vedaldi, Andrew Zisserman
Abstract	While semi-supervised learning (SSL) algorithms provide an efficient way to make use of both labelled and unlabelled data, they generally struggle when the number of annotated samples is very small. In this work, we consider the problem of SSL multi-class classification with very few labelled instances. We introduce two key ideas. The first is a simple but effective one: we leverage the power of transfer learning among different tasks and self-supervision to initialize a good representation of the data without making use of any label. The second idea is a new algorithm for SSL that can exploit well such a pre-trained representation. The algorithm works by alternating two phases, one fitting the labelled points and one fitting the unlabelled ones, with carefully-controlled information flow between them. The benefits are greatly reducing overfitting of the labelled data and avoiding issue with balancing labelled and unlabelled losses during training. We show empirically that this method can successfully train competitive models with as few as 10 labelled data points per class. More in general, we show that the idea of bootstrapping features using self-supervised learning always improves SSL on standard benchmarks. We show that our algorithm works increasingly well compared to other methods when refining from other tasks or datasets.
Tasks	Transfer Learning
Published	2019-05-21
URL	https://arxiv.org/abs/1905.08845v1
PDF	https://arxiv.org/pdf/1905.08845v1.pdf
PWC	https://paperswithcode.com/paper/semi-supervised-learning-with-scarce
Repo	https://github.com/srebuffi/semisup_scarse
Framework	pytorch

Model Comparison for Semantic Grouping


Title	Model Comparison for Semantic Grouping
Authors	Francisco Vargas, Kamen Brestnichki, Nils Hammerla
Abstract	We introduce a probabilistic framework for quantifying the semantic similarity between two groups of embeddings. We formulate the task of semantic similarity as a model comparison task in which we contrast a generative model which jointly models two sentences versus one that does not. We illustrate how this framework can be used for the Semantic Textual Similarity tasks using clear assumptions about how the embeddings of words are generated. We apply model comparison that utilises information criteria to address some of the shortcomings of Bayesian model comparison, whilst still penalising model complexity. We achieve competitive results by applying the proposed framework with an appropriate choice of likelihood on the STS datasets.
Tasks	Semantic Similarity, Semantic Textual Similarity
Published	2019-04-30
URL	http://arxiv.org/abs/1904.13323v2
PDF	http://arxiv.org/pdf/1904.13323v2.pdf
PWC	https://paperswithcode.com/paper/model-comparison-for-semantic-grouping
Repo	https://github.com/Babylonpartners/MCSG
Framework	pytorch

S4L: Self-Supervised Semi-Supervised Learning


Title	S4L: Self-Supervised Semi-Supervised Learning
Authors	Xiaohua Zhai, Avital Oliver, Alexander Kolesnikov, Lucas Beyer
Abstract	This work tackles the problem of semi-supervised learning of image classifiers. Our main insight is that the field of semi-supervised learning can benefit from the quickly advancing field of self-supervised visual representation learning. Unifying these two approaches, we propose the framework of self-supervised semi-supervised learning and use it to derive two novel semi-supervised image classification methods. We demonstrate the effectiveness of these methods in comparison to both carefully tuned baselines, and existing semi-supervised learning methods. We then show that our approach and existing semi-supervised methods can be jointly trained, yielding a new state-of-the-art result on semi-supervised ILSVRC-2012 with 10% of labels.
Tasks	Image Classification, Representation Learning, Semi-Supervised Image Classification
Published	2019-05-09
URL	https://arxiv.org/abs/1905.03670v2
PDF	https://arxiv.org/pdf/1905.03670v2.pdf
PWC	https://paperswithcode.com/paper/190503670
Repo	https://github.com/google-research/s4l
Framework	tf

Locality Constraint Dictionary Learning with Support Vector for Pattern Classification


Title	Locality Constraint Dictionary Learning with Support Vector for Pattern Classification
Authors	He-Feng Yin, Xiao-Jun Wu, Su-Gen Chen
Abstract	Discriminative dictionary learning (DDL) has recently gained significant attention due to its impressive performance in various pattern classification tasks. However, the locality of atoms is not fully explored in conventional DDL approaches which hampers their classification performance. In this paper, we propose a locality constraint dictionary learning with support vector discriminative term (LCDL-SV), in which the locality information is preserved by employing the graph Laplacian matrix of the learned dictionary. To jointly learn a classifier during the training phase, a support vector discriminative term is incorporated into the proposed objective function. Moreover, in the classification stage, the identity of test data is jointly determined by the regularized residual and the learned multi-class support vector machine. Finally, the resulting optimization problem is solved by utilizing the alternative strategy. Experimental results on benchmark databases demonstrate the superiority of our proposed method over previous dictionary learning approaches on both hand-crafted and deep features. The source code of our proposed LCDL-SV is accessible at https://github.com/yinhefeng/LCDL-SV
Tasks	Dictionary Learning
Published	2019-11-22
URL	https://arxiv.org/abs/1911.10003v1
PDF	https://arxiv.org/pdf/1911.10003v1.pdf
PWC	https://paperswithcode.com/paper/locality-constraint-dictionary-learning-with
Repo	https://github.com/yinhefeng/LCDL-SV
Framework	none

Image-and-Spatial Transformer Networks for Structure-Guided Image Registration


Title	Image-and-Spatial Transformer Networks for Structure-Guided Image Registration
Authors	Matthew C. H. Lee, Ozan Oktay, Andreas Schuh, Michiel Schaap, Ben Glocker
Abstract	Image registration with deep neural networks has become an active field of research and exciting avenue for a long standing problem in medical imaging. The goal is to learn a complex function that maps the appearance of input image pairs to parameters of a spatial transformation in order to align corresponding anatomical structures. We argue and show that the current direct, non-iterative approaches are sub-optimal, in particular if we seek accurate alignment of Structures-of-Interest (SoI). Information about SoI is often available at training time, for example, in form of segmentations or landmarks. We introduce a novel, generic framework, Image-and-Spatial Transformer Networks (ISTNs), to leverage SoI information allowing us to learn new image representations that are optimised for the downstream registration task. Thanks to these representations we can employ a test-specific, iterative refinement over the transformation parameters which yields highly accurate registration even with very limited training data. Performance is demonstrated on pairwise 3D brain registration and illustrative synthetic data.
Tasks	Image Registration
Published	2019-07-22
URL	https://arxiv.org/abs/1907.09200v1
PDF	https://arxiv.org/pdf/1907.09200v1.pdf
PWC	https://paperswithcode.com/paper/image-and-spatial-transformer-networks-for
Repo	https://github.com/biomedia-mira/istn
Framework	pytorch

Online matrix factorization for Markovian data and applications to Network Dictionary Learning


Title	Online matrix factorization for Markovian data and applications to Network Dictionary Learning
Authors	Hanbaek Lyu, Deanna Needell, Laura Balzano
Abstract	Online Matrix Factorization (OMF) is a fundamental tool for dictionary learning problems, giving an approximate representation of complex data sets in terms of a reduced number of extracted features. Convergence guarantees for most of the OMF algorithms in the literature assume independence between data matrices, and the case of a dependent data stream remains largely unexplored. In this paper, we show that the well-known OMF algorithm for i.i.d. stream of data proposed in \cite{mairal2010online}, in fact converges almost surely to the set of critical points of the expected loss function, even when the data matrices form a Markov chain satisfying a mild mixing condition. Furthermore, we extend the convergence result to the case when we can only approximately solve each step of the optimization problems in the algorithm. For applications, we demonstrate dictionary learning from a sequence of images generated by a Markov Chain Monte Carlo (MCMC) sampler. Lastly, by combining online non-negative matrix factorization and a recent MCMC algorithm for sampling motifs from networks, we propose a novel framework of Network Dictionary Learning, which extracts `network dictionary patches’ from a given network in an online manner that encodes main features of the network. We demonstrate this technique on real-world text data. \|
Tasks	Dictionary Learning
Published	2019-11-05
URL	https://arxiv.org/abs/1911.01931v3
PDF	https://arxiv.org/pdf/1911.01931v3.pdf
PWC	https://paperswithcode.com/paper/online-matrix-factorization-for-markovian
Repo	https://github.com/HanbaekLyu/ONMF_ONTF_NDL
Framework	none

Drone Shadow Tracking


Title	Drone Shadow Tracking
Authors	Xiaoyan Zou, Ruofan Zhou, Majed El Helou, Sabine Süsstrunk
Abstract	Aerial videos taken by a drone not too far above the surface may contain the drone’s shadow projected on the scene. This deteriorates the aesthetic quality of videos. With the presence of other shadows, shadow removal cannot be directly applied, and the shadow of the drone must be tracked. Tracking a drone’s shadow in a video is, however, challenging. The varying size, shape, change of orientation and drone altitude pose difficulties. The shadow can also easily disappear over dark areas. However, a shadow has specific properties that can be leveraged, besides its geometric shape. In this paper, we incorporate knowledge of the shadow’s physical properties, in the form of shadow detection masks, into a correlation-based tracking algorithm. We capture a test set of aerial videos taken with different settings and compare our results to those of a state-of-the-art tracking algorithm.
Tasks	Shadow Detection
Published	2019-05-20
URL	https://arxiv.org/abs/1905.08214v1
PDF	https://arxiv.org/pdf/1905.08214v1.pdf
PWC	https://paperswithcode.com/paper/drone-shadow-tracking
Repo	https://github.com/IVRL/Drone-Shadow-Tracking
Framework	none