July 29, 2019

2821 words 14 mins read

Paper Group AWR 152

Robust Adversarial Reinforcement Learning. Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery. Breast density classification with deep convolutional neural networks. Zero-Shot Relation Extraction via Reading Comprehension. DyNet: The Dynamic Neural Network Toolkit. An Optimization Approach to Learning Fall …

Robust Adversarial Reinforcement Learning


Title	Robust Adversarial Reinforcement Learning
Authors	Lerrel Pinto, James Davidson, Rahul Sukthankar, Abhinav Gupta
Abstract	Deep neural networks coupled with fast simulation and improved computation have led to recent successes in the field of reinforcement learning (RL). However, most current RL-based approaches fail to generalize since: (a) the gap between simulation and real world is so large that policy-learning approaches fail to transfer; (b) even if policy learning is done in real world, the data scarcity leads to failed generalization from training to test scenarios (e.g., due to different friction or object masses). Inspired from H-infinity control methods, we note that both modeling errors and differences in training and test scenarios can be viewed as extra forces/disturbances in the system. This paper proposes the idea of robust adversarial reinforcement learning (RARL), where we train an agent to operate in the presence of a destabilizing adversary that applies disturbance forces to the system. The jointly trained adversary is reinforced – that is, it learns an optimal destabilization policy. We formulate the policy learning as a zero-sum, minimax objective function. Extensive experiments in multiple environments (InvertedPendulum, HalfCheetah, Swimmer, Hopper and Walker2d) conclusively demonstrate that our method (a) improves training stability; (b) is robust to differences in training/test conditions; and c) outperform the baseline even in the absence of the adversary.
Tasks
Published	2017-03-08
URL	http://arxiv.org/abs/1703.02702v1
PDF	http://arxiv.org/pdf/1703.02702v1.pdf
PWC	https://paperswithcode.com/paper/robust-adversarial-reinforcement-learning
Repo	https://github.com/davidsonic/robust-grasp
Framework	tf

Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery


Title	Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery
Authors	Thomas Schlegl, Philipp Seeböck, Sebastian M. Waldstein, Ursula Schmidt-Erfurth, Georg Langs
Abstract	Obtaining models that capture imaging markers relevant for disease progression and treatment monitoring is challenging. Models are typically based on large amounts of data with annotated examples of known markers aiming at automating detection. High annotation effort and the limitation to a vocabulary of known markers limit the power of such approaches. Here, we perform unsupervised learning to identify anomalies in imaging data as candidates for markers. We propose AnoGAN, a deep convolutional generative adversarial network to learn a manifold of normal anatomical variability, accompanying a novel anomaly scoring scheme based on the mapping from image space to a latent space. Applied to new data, the model labels anomalies, and scores image patches indicating their fit into the learned distribution. Results on optical coherence tomography images of the retina demonstrate that the approach correctly identifies anomalous images, such as images containing retinal fluid or hyperreflective foci.
Tasks	Anomaly Detection, Unsupervised Anomaly Detection
Published	2017-03-17
URL	http://arxiv.org/abs/1703.05921v1
PDF	http://arxiv.org/pdf/1703.05921v1.pdf
PWC	https://paperswithcode.com/paper/unsupervised-anomaly-detection-with
Repo	https://github.com/seungjunlee96/DeepLearning-for-Medical-Imaging
Framework	none

Breast density classification with deep convolutional neural networks


Title	Breast density classification with deep convolutional neural networks
Authors	Nan Wu, Krzysztof J. Geras, Yiqiu Shen, Jingyi Su, S. Gene Kim, Eric Kim, Stacey Wolfson, Linda Moy, Kyunghyun Cho
Abstract	Breast density classification is an essential part of breast cancer screening. Although a lot of prior work considered this problem as a task for learning algorithms, to our knowledge, all of them used small and not clinically realistic data both for training and evaluation of their models. In this work, we explore the limits of this task with a data set coming from over 200,000 breast cancer screening exams. We use this data to train and evaluate a strong convolutional neural network classifier. In a reader study, we find that our model can perform this task comparably to a human expert.
Tasks	breast density classification
Published	2017-11-10
URL	http://arxiv.org/abs/1711.03674v1
PDF	http://arxiv.org/pdf/1711.03674v1.pdf
PWC	https://paperswithcode.com/paper/breast-density-classification-with-deep
Repo	https://github.com/nyukat/breast_density_classifier
Framework	pytorch

Zero-Shot Relation Extraction via Reading Comprehension


Title	Zero-Shot Relation Extraction via Reading Comprehension
Authors	Omer Levy, Minjoon Seo, Eunsol Choi, Luke Zettlemoyer
Abstract	We show that relation extraction can be reduced to answering simple reading comprehension questions, by associating one or more natural-language questions with each relation slot. This reduction has several advantages: we can (1) learn relation-extraction models by extending recent neural reading-comprehension techniques, (2) build very large training sets for those models by combining relation-specific crowd-sourced questions with distant supervision, and even (3) do zero-shot learning by extracting new relation types that are only specified at test-time, for which we have no labeled training examples. Experiments on a Wikipedia slot-filling task demonstrate that the approach can generalize to new questions for known relation types with high accuracy, and that zero-shot generalization to unseen relation types is possible, at lower accuracy levels, setting the bar for future work on this task.
Tasks	Reading Comprehension, Relation Extraction, Slot Filling, Zero-Shot Learning
Published	2017-06-13
URL	http://arxiv.org/abs/1706.04115v1
PDF	http://arxiv.org/pdf/1706.04115v1.pdf
PWC	https://paperswithcode.com/paper/zero-shot-relation-extraction-via-reading
Repo	https://github.com/zhuzhicai/SQuAD2.0-Baseline-Test-with-BiDAF-No-Answer
Framework	tf

DyNet: The Dynamic Neural Network Toolkit


Title	DyNet: The Dynamic Neural Network Toolkit
Authors	Graham Neubig, Chris Dyer, Yoav Goldberg, Austin Matthews, Waleed Ammar, Antonios Anastasopoulos, Miguel Ballesteros, David Chiang, Daniel Clothiaux, Trevor Cohn, Kevin Duh, Manaal Faruqui, Cynthia Gan, Dan Garrette, Yangfeng Ji, Lingpeng Kong, Adhiguna Kuncoro, Gaurav Kumar, Chaitanya Malaviya, Paul Michel, Yusuke Oda, Matthew Richardson, Naomi Saphra, Swabha Swayamdipta, Pengcheng Yin
Abstract	We describe DyNet, a toolkit for implementing neural network models based on dynamic declaration of network structure. In the static declaration strategy that is used in toolkits like Theano, CNTK, and TensorFlow, the user first defines a computation graph (a symbolic representation of the computation), and then examples are fed into an engine that executes this computation and computes its derivatives. In DyNet’s dynamic declaration strategy, computation graph construction is mostly transparent, being implicitly constructed by executing procedural code that computes the network outputs, and the user is free to use different network structures for each input. Dynamic declaration thus facilitates the implementation of more complicated network architectures, and DyNet is specifically designed to allow users to implement their models in a way that is idiomatic in their preferred programming language (C++ or Python). One challenge with dynamic declaration is that because the symbolic computation graph is defined anew for every training example, its construction must have low overhead. To achieve this, DyNet has an optimized C++ backend and lightweight graph representation. Experiments show that DyNet’s speeds are faster than or comparable with static declaration toolkits, and significantly faster than Chainer, another dynamic declaration toolkit. DyNet is released open-source under the Apache 2.0 license and available at http://github.com/clab/dynet.
Tasks	graph construction
Published	2017-01-15
URL	http://arxiv.org/abs/1701.03980v1
PDF	http://arxiv.org/pdf/1701.03980v1.pdf
PWC	https://paperswithcode.com/paper/dynet-the-dynamic-neural-network-toolkit
Repo	https://github.com/hanyaqian/dynet
Framework	none

An Optimization Approach to Learning Falling Rule Lists


Title	An Optimization Approach to Learning Falling Rule Lists
Authors	Chaofan Chen, Cynthia Rudin
Abstract	A falling rule list is a probabilistic decision list for binary classification, consisting of a series of if-then rules with antecedents in the if clauses and probabilities of the desired outcome (“1”) in the then clauses. Just as in a regular decision list, the order of rules in a falling rule list is important – each example is classified by the first rule whose antecedent it satisfies. Unlike a regular decision list, a falling rule list requires the probabilities of the desired outcome (“1”) to be monotonically decreasing down the list. We propose an optimization approach to learning falling rule lists and “softly” falling rule lists, along with Monte-Carlo search algorithms that use bounds on the optimal solution to prune the search space.
Tasks
Published	2017-10-06
URL	http://arxiv.org/abs/1710.02572v3
PDF	http://arxiv.org/pdf/1710.02572v3.pdf
PWC	https://paperswithcode.com/paper/an-optimization-approach-to-learning-falling
Repo	https://github.com/cfchen-duke/FRLOptimization
Framework	none

CNN features are also great at unsupervised classification


Title	CNN features are also great at unsupervised classification
Authors	Joris Guérin, Olivier Gibaru, Stéphane Thiery, Eric Nyiri
Abstract	This paper aims at providing insight on the transferability of deep CNN features to unsupervised problems. We study the impact of different pretrained CNN feature extractors on the problem of image set clustering for object classification as well as fine-grained classification. We propose a rather straightforward pipeline combining deep-feature extraction using a CNN pretrained on ImageNet and a classic clustering algorithm to classify sets of images. This approach is compared to state-of-the-art algorithms in image-clustering and provides better results. These results strengthen the belief that supervised training of deep CNN on large datasets, with a large variability of classes, extracts better features than most carefully designed engineering approaches, even for unsupervised tasks. We also validate our approach on a robotic application, consisting in sorting and storing objects smartly based on clustering.
Tasks	Image Clustering, Object Classification
Published	2017-07-06
URL	http://arxiv.org/abs/1707.01700v2
PDF	http://arxiv.org/pdf/1707.01700v2.pdf
PWC	https://paperswithcode.com/paper/cnn-features-are-also-great-at-unsupervised
Repo	https://github.com/jorisguerin/toolClustering_dataset
Framework	none

Referenceless Quality Estimation for Natural Language Generation


Title	Referenceless Quality Estimation for Natural Language Generation
Authors	Ondřej Dušek, Jekaterina Novikova, Verena Rieser
Abstract	Traditional automatic evaluation measures for natural language generation (NLG) use costly human-authored references to estimate the quality of a system output. In this paper, we propose a referenceless quality estimation (QE) approach based on recurrent neural networks, which predicts a quality score for a NLG system output by comparing it to the source meaning representation only. Our method outperforms traditional metrics and a constant baseline in most respects; we also show that synthetic data helps to increase correlation results by 21% compared to the base system. Our results are comparable to results obtained in similar QE tasks despite the more challenging setting.
Tasks	Text Generation
Published	2017-08-05
URL	http://arxiv.org/abs/1708.01759v1
PDF	http://arxiv.org/pdf/1708.01759v1.pdf
PWC	https://paperswithcode.com/paper/referenceless-quality-estimation-for-natural
Repo	https://github.com/tuetschek/ratpred
Framework	none

Learning Compact Geometric Features


Title	Learning Compact Geometric Features
Authors	Marc Khoury, Qian-Yi Zhou, Vladlen Koltun
Abstract	We present an approach to learning features that represent the local geometry around a point in an unstructured point cloud. Such features play a central role in geometric registration, which supports diverse applications in robotics and 3D vision. Current state-of-the-art local features for unstructured point clouds have been manually crafted and none combines the desirable properties of precision, compactness, and robustness. We show that features with these properties can be learned from data, by optimizing deep networks that map high-dimensional histograms into low-dimensional Euclidean spaces. The presented approach yields a family of features, parameterized by dimension, that are both more compact and more accurate than existing descriptors.
Tasks
Published	2017-09-15
URL	http://arxiv.org/abs/1709.05056v1
PDF	http://arxiv.org/pdf/1709.05056v1.pdf
PWC	https://paperswithcode.com/paper/learning-compact-geometric-features
Repo	https://github.com/marckhoury/CGF
Framework	none

Wasserstein Auto-Encoders


Title	Wasserstein Auto-Encoders
Authors	Ilya Tolstikhin, Olivier Bousquet, Sylvain Gelly, Bernhard Schoelkopf
Abstract	We propose the Wasserstein Auto-Encoder (WAE)—a new algorithm for building a generative model of the data distribution. WAE minimizes a penalized form of the Wasserstein distance between the model distribution and the target distribution, which leads to a different regularizer than the one used by the Variational Auto-Encoder (VAE). This regularizer encourages the encoded training distribution to match the prior. We compare our algorithm with several other techniques and show that it is a generalization of adversarial auto-encoders (AAE). Our experiments show that WAE shares many of the properties of VAEs (stable training, encoder-decoder architecture, nice latent manifold structure) while generating samples of better quality, as measured by the FID score.
Tasks
Published	2017-11-05
URL	https://arxiv.org/abs/1711.01558v4
PDF	https://arxiv.org/pdf/1711.01558v4.pdf
PWC	https://paperswithcode.com/paper/wasserstein-auto-encoders
Repo	https://github.com/pravn/wasserstein_autoencoders
Framework	pytorch

Fine-tuning Tree-LSTM for phrase-level sentiment classification on a Polish dependency treebank. Submission to PolEval task 2


Title	Fine-tuning Tree-LSTM for phrase-level sentiment classification on a Polish dependency treebank. Submission to PolEval task 2
Authors	Tomasz Korbak, Paulina Żak
Abstract	We describe a variant of Child-Sum Tree-LSTM deep neural network (Tai et al, 2015) fine-tuned for working with dependency trees and morphologically rich languages using the example of Polish. Fine-tuning included applying a custom regularization technique (zoneout, described by (Krueger et al., 2016), and further adapted for Tree-LSTMs) as well as using pre-trained word embeddings enhanced with sub-word information (Bojanowski et al., 2016). The system was implemented in PyTorch and evaluated on phrase-level sentiment labeling task as part of the PolEval competition.
Tasks	Sentiment Analysis, Word Embeddings
Published	2017-11-03
URL	http://arxiv.org/abs/1711.01985v1
PDF	http://arxiv.org/pdf/1711.01985v1.pdf
PWC	https://paperswithcode.com/paper/fine-tuning-tree-lstm-for-phrase-level
Repo	https://github.com/tomekkorbak/treehopper
Framework	pytorch

High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs


Title	High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
Authors	Ting-Chun Wang, Ming-Yu Liu, Jun-Yan Zhu, Andrew Tao, Jan Kautz, Bryan Catanzaro
Abstract	We present a new method for synthesizing high-resolution photo-realistic images from semantic label maps using conditional generative adversarial networks (conditional GANs). Conditional GANs have enabled a variety of applications, but the results are often limited to low-resolution and still far from realistic. In this work, we generate 2048x1024 visually appealing results with a novel adversarial loss, as well as new multi-scale generator and discriminator architectures. Furthermore, we extend our framework to interactive visual manipulation with two additional features. First, we incorporate object instance segmentation information, which enables object manipulations such as removing/adding objects and changing the object category. Second, we propose a method to generate diverse results given the same input, allowing users to edit the object appearance interactively. Human opinion studies demonstrate that our method significantly outperforms existing methods, advancing both the quality and the resolution of deep image synthesis and editing.
Tasks	Conditional Image Generation, Image Generation, Image-to-Image Translation, Instance Segmentation, Semantic Segmentation
Published	2017-11-30
URL	http://arxiv.org/abs/1711.11585v2
PDF	http://arxiv.org/pdf/1711.11585v2.pdf
PWC	https://paperswithcode.com/paper/high-resolution-image-synthesis-and-semantic
Repo	https://github.com/eyeswideopen/pix2pixHD-MED
Framework	pytorch

Statistical Parametric Speech Synthesis Incorporating Generative Adversarial Networks


Title	Statistical Parametric Speech Synthesis Incorporating Generative Adversarial Networks
Authors	Yuki Saito, Shinnosuke Takamichi, Hiroshi Saruwatari
Abstract	A method for statistical parametric speech synthesis incorporating generative adversarial networks (GANs) is proposed. Although powerful deep neural networks (DNNs) techniques can be applied to artificially synthesize speech waveform, the synthetic speech quality is low compared with that of natural speech. One of the issues causing the quality degradation is an over-smoothing effect often observed in the generated speech parameters. A GAN introduced in this paper consists of two neural networks: a discriminator to distinguish natural and generated samples, and a generator to deceive the discriminator. In the proposed framework incorporating the GANs, the discriminator is trained to distinguish natural and generated speech parameters, while the acoustic models are trained to minimize the weighted sum of the conventional minimum generation loss and an adversarial loss for deceiving the discriminator. Since the objective of the GANs is to minimize the divergence (i.e., distribution difference) between the natural and generated speech parameters, the proposed method effectively alleviates the over-smoothing effect on the generated speech parameters. We evaluated the effectiveness for text-to-speech and voice conversion, and found that the proposed method can generate more natural spectral parameters and $F_0$ than conventional minimum generation error training algorithm regardless its hyper-parameter settings. Furthermore, we investigated the effect of the divergence of various GANs, and found that a Wasserstein GAN minimizing the Earth-Mover’s distance works the best in terms of improving synthetic speech quality.
Tasks	Speech Synthesis, Voice Conversion
Published	2017-09-23
URL	http://arxiv.org/abs/1709.08041v1
PDF	http://arxiv.org/pdf/1709.08041v1.pdf
PWC	https://paperswithcode.com/paper/statistical-parametric-speech-synthesis-1
Repo	https://github.com/rickyHong/GAN-TTS-repl2
Framework	pytorch

DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networks


Title	DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networks
Authors	Sen Wang, Ronald Clark, Hongkai Wen, Niki Trigoni
Abstract	This paper studies monocular visual odometry (VO) problem. Most of existing VO algorithms are developed under a standard pipeline including feature extraction, feature matching, motion estimation, local optimisation, etc. Although some of them have demonstrated superior performance, they usually need to be carefully designed and specifically fine-tuned to work well in different environments. Some prior knowledge is also required to recover an absolute scale for monocular VO. This paper presents a novel end-to-end framework for monocular VO by using deep Recurrent Convolutional Neural Networks (RCNNs). Since it is trained and deployed in an end-to-end manner, it infers poses directly from a sequence of raw RGB images (videos) without adopting any module in the conventional VO pipeline. Based on the RCNNs, it not only automatically learns effective feature representation for the VO problem through Convolutional Neural Networks, but also implicitly models sequential dynamics and relations using deep Recurrent Neural Networks. Extensive experiments on the KITTI VO dataset show competitive performance to state-of-the-art methods, verifying that the end-to-end Deep Learning technique can be a viable complement to the traditional VO systems.
Tasks	Monocular Visual Odometry, Motion Estimation, Visual Odometry
Published	2017-09-25
URL	http://arxiv.org/abs/1709.08429v1
PDF	http://arxiv.org/pdf/1709.08429v1.pdf
PWC	https://paperswithcode.com/paper/deepvo-towards-end-to-end-visual-odometry
Repo	https://github.com/fshamshirdar/DeepVO
Framework	pytorch

Construction of a Japanese Word Similarity Dataset


Title	Construction of a Japanese Word Similarity Dataset
Authors	Yuya Sakaizawa, Mamoru Komachi
Abstract	An evaluation of distributed word representation is generally conducted using a word similarity task and/or a word analogy task. There are many datasets readily available for these tasks in English. However, evaluating distributed representation in languages that do not have such resources (e.g., Japanese) is difficult. Therefore, as a first step toward evaluating distributed representations in Japanese, we constructed a Japanese word similarity dataset. To the best of our knowledge, our dataset is the first resource that can be used to evaluate distributed representations in Japanese. Moreover, our dataset contains various parts of speech and includes rare words in addition to common words.
Tasks
Published	2017-03-17
URL	http://arxiv.org/abs/1703.05916v2
PDF	http://arxiv.org/pdf/1703.05916v2.pdf
PWC	https://paperswithcode.com/paper/construction-of-a-japanese-word-similarity
Repo	https://github.com/kdrl/SCNE
Framework	none