October 20, 2019

3174 words 15 mins read

Paper Group AWR 337

Sampled Policy Gradient for Learning to Play the Game Agar.io. Exploring object-centric and scene-centric CNN features and their complementarity for human rights violations recognition in images. Effective Subword Segmentation for Text Comprehension. Learning for Video Super-Resolution through HR Optical Flow Estimation. Bias and Generalization in …

Sampled Policy Gradient for Learning to Play the Game Agar.io


Title	Sampled Policy Gradient for Learning to Play the Game Agar.io
Authors	Anton Orell Wiehe, Nil Stolt Ansó, Madalina M. Drugan, Marco A. Wiering
Abstract	In this paper, a new offline actor-critic learning algorithm is introduced: Sampled Policy Gradient (SPG). SPG samples in the action space to calculate an approximated policy gradient by using the critic to evaluate the samples. This sampling allows SPG to search the action-Q-value space more globally than deterministic policy gradient (DPG), enabling it to theoretically avoid more local optima. SPG is compared to Q-learning and the actor-critic algorithms CACLA and DPG in a pellet collection task and a self play environment in the game Agar.io. The online game Agar.io has become massively popular on the internet due to intuitive game design and the ability to instantly compete against players around the world. From the point of view of artificial intelligence this game is also very intriguing: The game has a continuous input and action space and allows to have diverse agents with complex strategies compete against each other. The experimental results show that Q-Learning and CACLA outperform a pre-programmed greedy bot in the pellet collection task, but all algorithms fail to outperform this bot in a fighting scenario. The SPG algorithm is analyzed to have great extendability through offline exploration and it matches DPG in performance even in its basic form without extensive sampling.
Tasks	Q-Learning
Published	2018-09-15
URL	http://arxiv.org/abs/1809.05763v1
PDF	http://arxiv.org/pdf/1809.05763v1.pdf
PWC	https://paperswithcode.com/paper/sampled-policy-gradient-for-learning-to-play
Repo	https://github.com/RUKip/MachineLearningProject
Framework	none

Exploring object-centric and scene-centric CNN features and their complementarity for human rights violations recognition in images


Title	Exploring object-centric and scene-centric CNN features and their complementarity for human rights violations recognition in images
Authors	Grigorios Kalliatakis, Shoaib Ehsan, Ales Leonardis, Klaus McDonald-Maier
Abstract	Identifying potential abuses of human rights through imagery is a novel and challenging task in the field of computer vision, that will enable to expose human rights violations over large-scale data that may otherwise be impossible. While standard databases for object and scene categorisation contain hundreds of different classes, the largest available dataset of human rights violations contains only 4 classes. Here, we introduce the `Human Rights Archive Database’ (HRA), a verified-by-experts repository of 3050 human rights violations photographs, labelled with human rights semantic categories, comprising a list of the types of human rights abuses encountered at present. With the HRA dataset and a two-phase transfer learning scheme, we fine-tuned the state-of-the-art deep convolutional neural networks (CNNs) to provide human rights violations classification CNNs (HRA-CNNs). We also present extensive experiments refined to evaluate how well object-centric and scene-centric CNN features can be combined for the task of recognising human rights abuses. With this, we show that HRA database poses a challenge at a higher level for the well studied representation learning methods, and provide a benchmark in the task of human rights violations recognition in visual context. We expect this dataset can help to open up new horizons on creating systems able of recognising rich information about human rights violations. Our dataset, codes and trained models are available online at https://github.com/GKalliatakis/Human-Rights-Archive-CNNs. \|
Tasks	Representation Learning, Transfer Learning
Published	2018-05-12
URL	http://arxiv.org/abs/1805.04714v1
PDF	http://arxiv.org/pdf/1805.04714v1.pdf
PWC	https://paperswithcode.com/paper/exploring-object-centric-and-scene-centric
Repo	https://github.com/GKalliatakis/Human-Rights-Archive-CNNs
Framework	tf

Effective Subword Segmentation for Text Comprehension


Title	Effective Subword Segmentation for Text Comprehension
Authors	Zhuosheng Zhang, Hai Zhao, Kangwei Ling, Jiangtong Li, Zuchao Li, Shexia He, Guohong Fu
Abstract	Representation learning is the foundation of machine reading comprehension and inference. In state-of-the-art models, character-level representations have been broadly adopted to alleviate the problem of effectively representing rare or complex words. However, character itself is not a natural minimal linguistic unit for representation or word embedding composing due to ignoring the linguistic coherence of consecutive characters inside word. This paper presents a general subword-augmented embedding framework for learning and composing computationally-derived subword-level representations. We survey a series of unsupervised segmentation methods for subword acquisition and different subword-augmented strategies for text understanding, showing that subword-augmented embedding significantly improves our baselines in various types of text understanding tasks on both English and Chinese benchmarks.
Tasks	Machine Reading Comprehension, Reading Comprehension, Representation Learning
Published	2018-11-06
URL	https://arxiv.org/abs/1811.02364v2
PDF	https://arxiv.org/pdf/1811.02364v2.pdf
PWC	https://paperswithcode.com/paper/effective-subword-segmentation-for-text
Repo	https://github.com/cooelf/subword_seg
Framework	none

Learning for Video Super-Resolution through HR Optical Flow Estimation


Title	Learning for Video Super-Resolution through HR Optical Flow Estimation
Authors	Longguang Wang, Yulan Guo, Zaiping Lin, Xinpu Deng, Wei An
Abstract	Video super-resolution (SR) aims to generate a sequence of high-resolution (HR) frames with plausible and temporally consistent details from their low-resolution (LR) counterparts. The generation of accurate correspondence plays a significant role in video SR. It is demonstrated by traditional video SR methods that simultaneous SR of both images and optical flows can provide accurate correspondences and better SR results. However, LR optical flows are used in existing deep learning based methods for correspondence generation. In this paper, we propose an end-to-end trainable video SR framework to super-resolve both images and optical flows. Specifically, we first propose an optical flow reconstruction network (OFRnet) to infer HR optical flows in a coarse-to-fine manner. Then, motion compensation is performed according to the HR optical flows. Finally, compensated LR inputs are fed to a super-resolution network (SRnet) to generate the SR results. Extensive experiments demonstrate that HR optical flows provide more accurate correspondences than their LR counterparts and improve both accuracy and consistency performance. Comparative results on the Vid4 and DAVIS-10 datasets show that our framework achieves the state-of-the-art performance.
Tasks	Motion Compensation, Optical Flow Estimation, Super-Resolution, Video Super-Resolution
Published	2018-09-23
URL	http://arxiv.org/abs/1809.08573v2
PDF	http://arxiv.org/pdf/1809.08573v2.pdf
PWC	https://paperswithcode.com/paper/learning-for-video-super-resolution-through
Repo	https://github.com/LongguangWang/SOF-VSR
Framework	pytorch

Bias and Generalization in Deep Generative Models: An Empirical Study


Title	Bias and Generalization in Deep Generative Models: An Empirical Study
Authors	Shengjia Zhao, Hongyu Ren, Arianna Yuan, Jiaming Song, Noah Goodman, Stefano Ermon
Abstract	In high dimensional settings, density estimation algorithms rely crucially on their inductive bias. Despite recent empirical success, the inductive bias of deep generative models is not well understood. In this paper we propose a framework to systematically investigate bias and generalization in deep generative models of images. Inspired by experimental methods from cognitive psychology, we probe each learning algorithm with carefully designed training datasets to characterize when and how existing models generate novel attributes and their combinations. We identify similarities to human psychology and verify that these patterns are consistent across commonly used models and architectures.
Tasks	Density Estimation
Published	2018-11-08
URL	http://arxiv.org/abs/1811.03259v1
PDF	http://arxiv.org/pdf/1811.03259v1.pdf
PWC	https://paperswithcode.com/paper/bias-and-generalization-in-deep-generative
Repo	https://github.com/ermongroup/BiasAndGeneralization
Framework	tf

A Visual Attention Grounding Neural Model for Multimodal Machine Translation


Title	A Visual Attention Grounding Neural Model for Multimodal Machine Translation
Authors	Mingyang Zhou, Runxiang Cheng, Yong Jae Lee, Zhou Yu
Abstract	We introduce a novel multimodal machine translation model that utilizes parallel visual and textual information. Our model jointly optimizes the learning of a shared visual-language embedding and a translator. The model leverages a visual attention grounding mechanism that links the visual semantics with the corresponding textual semantics. Our approach achieves competitive state-of-the-art results on the Multi30K and the Ambiguous COCO datasets. We also collected a new multilingual multimodal product description dataset to simulate a real-world international online shopping scenario. On this dataset, our visual attention grounding model outperforms other methods by a large margin.
Tasks	Machine Translation, Multimodal Machine Translation
Published	2018-08-24
URL	http://arxiv.org/abs/1808.08266v2
PDF	http://arxiv.org/pdf/1808.08266v2.pdf
PWC	https://paperswithcode.com/paper/a-visual-attention-grounding-neural-model-for
Repo	https://github.com/sampalomad/IKEA-Dataset
Framework	none

One-Shot Relational Learning for Knowledge Graphs


Title	One-Shot Relational Learning for Knowledge Graphs
Authors	Wenhan Xiong, Mo Yu, Shiyu Chang, Xiaoxiao Guo, William Yang Wang
Abstract	Knowledge graphs (KGs) are the key components of various natural language processing applications. To further expand KGs’ coverage, previous studies on knowledge graph completion usually require a large number of training instances for each relation. However, we observe that long-tail relations are actually more common in KGs and those newly added relations often do not have many known triples for training. In this work, we aim at predicting new facts under a challenging setting where only one training instance is available. We propose a one-shot relational learning framework, which utilizes the knowledge extracted by embedding models and learns a matching metric by considering both the learned embeddings and one-hop graph structures. Empirically, our model yields considerable performance improvements over existing embedding models, and also eliminates the need of re-training the embedding models when dealing with newly added relations.
Tasks	Knowledge Graph Completion, Knowledge Graphs, Relational Reasoning
Published	2018-08-27
URL	http://arxiv.org/abs/1808.09040v1
PDF	http://arxiv.org/pdf/1808.09040v1.pdf
PWC	https://paperswithcode.com/paper/one-shot-relational-learning-for-knowledge
Repo	https://github.com/xwhan/One-shot-Relational-Learning
Framework	pytorch

Solving Statistical Mechanics Using Variational Autoregressive Networks


Title	Solving Statistical Mechanics Using Variational Autoregressive Networks
Authors	Dian Wu, Lei Wang, Pan Zhang
Abstract	We propose a general framework for solving statistical mechanics of systems with finite size. The approach extends the celebrated variational mean-field approaches using autoregressive neural networks, which support direct sampling and exact calculation of normalized probability of configurations. It computes variational free energy, estimates physical quantities such as entropy, magnetizations and correlations, and generates uncorrelated samples all at once. Training of the network employs the policy gradient approach in reinforcement learning, which unbiasedly estimates the gradient of variational parameters. We apply our approach to several classic systems, including 2D Ising models, the Hopfield model, the Sherrington-Kirkpatrick model, and the inverse Ising model, for demonstrating its advantages over existing variational mean-field methods. Our approach sheds light on solving statistical physics problems using modern deep generative neural networks.
Tasks
Published	2018-09-27
URL	http://arxiv.org/abs/1809.10606v2
PDF	http://arxiv.org/pdf/1809.10606v2.pdf
PWC	https://paperswithcode.com/paper/solving-statistical-mechanics-using
Repo	https://github.com/wdphy16/stat-mech-van
Framework	pytorch

Unsupervised Feature Learning of Human Actions as Trajectories in Pose Embedding Manifold


Title	Unsupervised Feature Learning of Human Actions as Trajectories in Pose Embedding Manifold
Authors	Jogendra Nath Kundu, Maharshi Gor, Phani Krishna Uppala, R. Venkatesh Babu
Abstract	An unsupervised human action modeling framework can provide useful pose-sequence representation, which can be utilized in a variety of pose analysis applications. In this work we propose a novel temporal pose-sequence modeling framework, which can embed the dynamics of 3D human-skeleton joints to a continuous latent space in an efficient manner. In contrast to end-to-end framework explored by previous works, we disentangle the task of individual pose representation learning from the task of learning actions as a trajectory in pose embedding space. In order to realize a continuous pose embedding manifold with improved reconstructions, we propose an unsupervised, manifold learning procedure named Encoder GAN, (or EnGAN). Further, we use the pose embeddings generated by EnGAN to model human actions using a bidirectional RNN auto-encoder architecture, PoseRNN. We introduce first-order gradient loss to explicitly enforce temporal regularity in the predicted motion sequence. A hierarchical feature fusion technique is also investigated for simultaneous modeling of local skeleton joints along with global pose variations. We demonstrate state-of-the-art transfer-ability of the learned representation against other supervisedly and unsupervisedly learned motion embeddings for the task of fine-grained action recognition on SBU interaction dataset. Further, we show the qualitative strengths of the proposed framework by visualizing skeleton pose reconstructions and interpolations in pose-embedding space, and low dimensional principal component projections of the reconstructed pose trajectories.
Tasks	Representation Learning, Temporal Action Localization
Published	2018-12-06
URL	http://arxiv.org/abs/1812.02592v1
PDF	http://arxiv.org/pdf/1812.02592v1.pdf
PWC	https://paperswithcode.com/paper/unsupervised-feature-learning-of-human
Repo	https://github.com/maharshi95/Pose2vec
Framework	tf

Autofocus Layer for Semantic Segmentation


Title	Autofocus Layer for Semantic Segmentation
Authors	Yao Qin, Konstantinos Kamnitsas, Siddharth Ancha, Jay Nanavati, Garrison Cottrell, Antonio Criminisi, Aditya Nori
Abstract	We propose the autofocus convolutional layer for semantic segmentation with the objective of enhancing the capabilities of neural networks for multi-scale processing. Autofocus layers adaptively change the size of the effective receptive field based on the processed context to generate more powerful features. This is achieved by parallelising multiple convolutional layers with different dilation rates, combined by an attention mechanism that learns to focus on the optimal scales driven by context. By sharing the weights of the parallel convolutions we make the network scale-invariant, with only a modest increase in the number of parameters. The proposed autofocus layer can be easily integrated into existing networks to improve a model’s representational power. We evaluate our models on the challenging tasks of multi-organ segmentation in pelvic CT and brain tumor segmentation in MRI and achieve very promising performance.
Tasks	Brain Tumor Segmentation, Medical Image Segmentation, Semantic Segmentation
Published	2018-05-22
URL	http://arxiv.org/abs/1805.08403v3
PDF	http://arxiv.org/pdf/1805.08403v3.pdf
PWC	https://paperswithcode.com/paper/autofocus-layer-for-semantic-segmentation
Repo	https://github.com/perslev/Autofocus-Layer-TF
Framework	tf

Solving Imperfect-Information Games via Discounted Regret Minimization


Title	Solving Imperfect-Information Games via Discounted Regret Minimization
Authors	Noam Brown, Tuomas Sandholm
Abstract	Counterfactual regret minimization (CFR) is a family of iterative algorithms that are the most popular and, in practice, fastest approach to approximately solving large imperfect-information games. In this paper we introduce novel CFR variants that 1) discount regrets from earlier iterations in various ways (in some cases differently for positive and negative regrets), 2) reweight iterations in various ways to obtain the output strategies, 3) use a non-standard regret minimizer and/or 4) leverage “optimistic regret matching”. They lead to dramatically improved performance in many settings. For one, we introduce a variant that outperforms CFR+, the prior state-of-the-art algorithm, in every game tested, including large-scale realistic settings. CFR+ is a formidable benchmark: no other algorithm has been able to outperform it. Finally, we show that, unlike CFR+, many of the important new variants are compatible with modern imperfect-information-game pruning techniques and one is also compatible with sampling in the game tree.
Tasks
Published	2018-09-11
URL	http://arxiv.org/abs/1809.04040v3
PDF	http://arxiv.org/pdf/1809.04040v3.pdf
PWC	https://paperswithcode.com/paper/solving-imperfect-information-games-via
Repo	https://github.com/maxschorer/ticket_to_ride
Framework	none

TensorFuzz: Debugging Neural Networks with Coverage-Guided Fuzzing


Title	TensorFuzz: Debugging Neural Networks with Coverage-Guided Fuzzing
Authors	Augustus Odena, Ian Goodfellow
Abstract	Machine learning models are notoriously difficult to interpret and debug. This is particularly true of neural networks. In this work, we introduce automated software testing techniques for neural networks that are well-suited to discovering errors which occur only for rare inputs. Specifically, we develop coverage-guided fuzzing (CGF) methods for neural networks. In CGF, random mutations of inputs to a neural network are guided by a coverage metric toward the goal of satisfying user-specified constraints. We describe how fast approximate nearest neighbor algorithms can provide this coverage metric. We then discuss the application of CGF to the following goals: finding numerical errors in trained neural networks, generating disagreements between neural networks and quantized versions of those networks, and surfacing undesirable behavior in character level language models. Finally, we release an open source library called TensorFuzz that implements the described techniques.
Tasks
Published	2018-07-28
URL	http://arxiv.org/abs/1807.10875v1
PDF	http://arxiv.org/pdf/1807.10875v1.pdf
PWC	https://paperswithcode.com/paper/tensorfuzz-debugging-neural-networks-with
Repo	https://github.com/sandeep-krishnamurthy/my-reading-list
Framework	mxnet

Deep autoregressive neural networks for high-dimensional inverse problems in groundwater contaminant source identification


Title	Deep autoregressive neural networks for high-dimensional inverse problems in groundwater contaminant source identification
Authors	Shaoxing Mo, Nicholas Zabaras, Xiaoqing Shi, Jichun Wu
Abstract	Identification of a groundwater contaminant source simultaneously with the hydraulic conductivity in highly-heterogeneous media often results in a high-dimensional inverse problem. In this study, a deep autoregressive neural network-based surrogate method is developed for the forward model to allow us to solve efficiently such high-dimensional inverse problems. The surrogate is trained using limited evaluations of the forward model. Since the relationship between the time-varying inputs and outputs of the forward transport model is complex, we propose an autoregressive strategy, which treats the output at the previous time step as input to the network for predicting the output at the current time step. We employ a dense convolutional encoder-decoder network architecture in which the high-dimensional input and output fields of the model are treated as images to leverage the robust capability of convolutional networks in image-like data processing. An iterative local updating ensemble smoother (ILUES) algorithm is used as the inversion framework. The proposed method is evaluated using a synthetic contaminant source identification problem with 686 uncertain input parameters. Results indicate that, with relatively limited training data, the deep autoregressive neural network consisting of 27 convolutional layers is capable of providing an accurate approximation for the high-dimensional model input-output relationship. The autoregressive strategy substantially improves the network’s accuracy and computational efficiency. The application of the surrogate-based ILUES in solving the inverse problem shows that it can achieve accurate inversion results and predictive uncertainty estimates.
Tasks
Published	2018-12-22
URL	http://arxiv.org/abs/1812.09444v1
PDF	http://arxiv.org/pdf/1812.09444v1.pdf
PWC	https://paperswithcode.com/paper/deep-autoregressive-neural-networks-for-high
Repo	https://github.com/cics-nd/cnn-inversion
Framework	pytorch

Evaluation of sentence embeddings in downstream and linguistic probing tasks


Title	Evaluation of sentence embeddings in downstream and linguistic probing tasks
Authors	Christian S. Perone, Roberto Silveira, Thomas S. Paula
Abstract	Despite the fast developmental pace of new sentence embedding methods, it is still challenging to find comprehensive evaluations of these different techniques. In the past years, we saw significant improvements in the field of sentence embeddings and especially towards the development of universal sentence encoders that could provide inductive transfer to a wide variety of downstream tasks. In this work, we perform a comprehensive evaluation of recent methods using a wide variety of downstream and linguistic feature probing tasks. We show that a simple approach using bag-of-words with a recently introduced language model for deep context-dependent word embeddings proved to yield better results in many tasks when compared to sentence encoders trained on entailment datasets. We also show, however, that we are still far away from a universal encoder that can perform consistently across several downstream tasks.
Tasks	Language Modelling, Sentence Embedding, Sentence Embeddings, Word Embeddings
Published	2018-06-16
URL	http://arxiv.org/abs/1806.06259v1
PDF	http://arxiv.org/pdf/1806.06259v1.pdf
PWC	https://paperswithcode.com/paper/evaluation-of-sentence-embeddings-in
Repo	https://github.com/cheng18/bilm-tf
Framework	tf

Probabilistic Random Forest: A machine learning algorithm for noisy datasets


Title	Probabilistic Random Forest: A machine learning algorithm for noisy datasets
Authors	Itamar Reis, Dalya Baron, Sahar Shahaf
Abstract	Machine learning (ML) algorithms become increasingly important in the analysis of astronomical data. However, since most ML algorithms are not designed to take data uncertainties into account, ML based studies are mostly restricted to data with high signal-to-noise ratio. Astronomical datasets of such high-quality are uncommon. In this work we modify the long-established Random Forest (RF) algorithm to take into account uncertainties in the measurements (i.e., features) as well as in the assigned classes (i.e., labels). To do so, the Probabilistic Random Forest (PRF) algorithm treats the features and labels as probability distribution functions, rather than deterministic quantities. We perform a variety of experiments where we inject different types of noise to a dataset, and compare the accuracy of the PRF to that of RF. The PRF outperforms RF in all cases, with a moderate increase in running time. We find an improvement in classification accuracy of up to 10% in the case of noisy features, and up to 30% in the case of noisy labels. The PRF accuracy decreased by less then 5% for a dataset with as many as 45% misclassified objects, compared to a clean dataset. Apart from improving the prediction accuracy in noisy datasets, the PRF naturally copes with missing values in the data, and outperforms RF when applied to a dataset with different noise characteristics in the training and test sets, suggesting that it can be used for Transfer Learning.
Tasks	Transfer Learning
Published	2018-11-14
URL	http://arxiv.org/abs/1811.05994v1
PDF	http://arxiv.org/pdf/1811.05994v1.pdf
PWC	https://paperswithcode.com/paper/probabilistic-random-forest-a-machine
Repo	https://github.com/ireis/PRF
Framework	none