July 27, 2019

3304 words 16 mins read

Paper Group ANR 660

Robust Task Clustering for Deep Many-Task Learning. Emergent Translation in Multi-Agent Communication. Can GAN Learn Topological Features of a Graph?. ACtuAL: Actor-Critic Under Adversarial Learning. Prediction Scores as a Window into Classifier Behavior. Dynamic Safe Interruptibility for Decentralized Multi-Agent Reinforcement Learning. I2T2I: Lea …

Robust Task Clustering for Deep Many-Task Learning


Title	Robust Task Clustering for Deep Many-Task Learning
Authors	Mo Yu, Xiaoxiao Guo, Jinfeng Yi, Shiyu Chang, Saloni Potdar, Gerald Tesauro, Haoyu Wang, Bowen Zhou
Abstract	We investigate task clustering for deep-learning based multi-task and few-shot learning in a many-task setting. We propose a new method to measure task similarities with cross-task transfer performance matrix for the deep learning scenario. Although this matrix provides us critical information regarding similarity between tasks, its asymmetric property and unreliable performance scores can affect conventional clustering methods adversely. Additionally, the uncertain task-pairs, i.e., the ones with extremely asymmetric transfer scores, may collectively mislead clustering algorithms to output an inaccurate task-partition. To overcome these limitations, we propose a novel task-clustering algorithm by using the matrix completion technique. The proposed algorithm constructs a partially-observed similarity matrix based on the certainty of cluster membership of the task-pairs. We then use a matrix completion algorithm to complete the similarity matrix. Our theoretical analysis shows that under mild constraints, the proposed algorithm will perfectly recover the underlying “true” similarity matrix with a high probability. Our results show that the new task clustering method can discover task clusters for training flexible and superior neural network models in a multi-task learning setup for sentiment classification and dialog intent classification tasks. Our task clustering approach also extends metric-based few-shot learning methods to adapt multiple metrics, which demonstrates empirical advantages when the tasks are diverse.
Tasks	Few-Shot Learning, Intent Classification, Matrix Completion, Multi-Task Learning, Sentiment Analysis
Published	2017-08-26
URL	http://arxiv.org/abs/1708.07918v2
PDF	http://arxiv.org/pdf/1708.07918v2.pdf
PWC	https://paperswithcode.com/paper/robust-task-clustering-for-deep-many-task
Repo
Framework

Emergent Translation in Multi-Agent Communication


Title	Emergent Translation in Multi-Agent Communication
Authors	Jason Lee, Kyunghyun Cho, Jason Weston, Douwe Kiela
Abstract	While most machine translation systems to date are trained on large parallel corpora, humans learn language in a different way: by being grounded in an environment and interacting with other humans. In this work, we propose a communication game where two agents, native speakers of their own respective languages, jointly learn to solve a visual referential task. We find that the ability to understand and translate a foreign language emerges as a means to achieve shared goals. The emergent translation is interactive and multimodal, and crucially does not require parallel corpora, but only monolingual, independent text and corresponding images. Our proposed translation model achieves this by grounding the source and target languages into a shared visual modality, and outperforms several baselines on both word-level and sentence-level translation tasks. Furthermore, we show that agents in a multilingual community learn to translate better and faster than in a bilingual communication setting.
Tasks	Machine Translation
Published	2017-10-12
URL	http://arxiv.org/abs/1710.06922v2
PDF	http://arxiv.org/pdf/1710.06922v2.pdf
PWC	https://paperswithcode.com/paper/emergent-translation-in-multi-agent
Repo
Framework

Can GAN Learn Topological Features of a Graph?


Title	Can GAN Learn Topological Features of a Graph?
Authors	Weiyi Liu, Pin-Yu Chen, Hal Cooper, Min Hwan Oh, Sailung Yeung, Toyotaro Suzumura
Abstract	This paper is first-line research expanding GANs into graph topology analysis. By leveraging the hierarchical connectivity structure of a graph, we have demonstrated that generative adversarial networks (GANs) can successfully capture topological features of any arbitrary graph, and rank edge sets by different stages according to their contribution to topology reconstruction. Moreover, in addition to acting as an indicator of graph reconstruction, we find that these stages can also preserve important topological features in a graph.
Tasks
Published	2017-07-19
URL	http://arxiv.org/abs/1707.06197v1
PDF	http://arxiv.org/pdf/1707.06197v1.pdf
PWC	https://paperswithcode.com/paper/can-gan-learn-topological-features-of-a-graph
Repo
Framework

ACtuAL: Actor-Critic Under Adversarial Learning


Title	ACtuAL: Actor-Critic Under Adversarial Learning
Authors	Anirudh Goyal, Nan Rosemary Ke, Alex Lamb, R Devon Hjelm, Chris Pal, Joelle Pineau, Yoshua Bengio
Abstract	Generative Adversarial Networks (GANs) are a powerful framework for deep generative modeling. Posed as a two-player minimax problem, GANs are typically trained end-to-end on real-valued data and can be used to train a generator of high-dimensional and realistic images. However, a major limitation of GANs is that training relies on passing gradients from the discriminator through the generator via back-propagation. This makes it fundamentally difficult to train GANs with discrete data, as generation in this case typically involves a non-differentiable function. These difficulties extend to the reinforcement learning setting when the action space is composed of discrete decisions. We address these issues by reframing the GAN framework so that the generator is no longer trained using gradients through the discriminator, but is instead trained using a learned critic in the actor-critic framework with a Temporal Difference (TD) objective. This is a natural fit for sequence modeling and we use it to achieve improvements on language modeling tasks over the standard Teacher-Forcing methods.
Tasks	Language Modelling
Published	2017-11-13
URL	http://arxiv.org/abs/1711.04755v1
PDF	http://arxiv.org/pdf/1711.04755v1.pdf
PWC	https://paperswithcode.com/paper/actual-actor-critic-under-adversarial
Repo
Framework

Prediction Scores as a Window into Classifier Behavior


Title	Prediction Scores as a Window into Classifier Behavior
Authors	Medha Katehara, Emma Beauxis-Aussalet, Bilal Alsallakh
Abstract	Most multi-class classifiers make their prediction for a test sample by scoring the classes and selecting the one with the highest score. Analyzing these prediction scores is useful to understand the classifier behavior and to assess its reliability. We present an interactive visualization that facilitates per-class analysis of these scores. Our system, called Classilist, enables relating these scores to the classification correctness and to the underlying samples and their features. We illustrate how such analysis reveals varying behavior of different classifiers. Classilist is available for use online, along with source code, video tutorials, and plugins for R, RapidMiner, and KNIME at https://katehara.github.io/classilist-site/.
Tasks
Published	2017-11-18
URL	http://arxiv.org/abs/1711.06795v1
PDF	http://arxiv.org/pdf/1711.06795v1.pdf
PWC	https://paperswithcode.com/paper/prediction-scores-as-a-window-into-classifier
Repo
Framework

Dynamic Safe Interruptibility for Decentralized Multi-Agent Reinforcement Learning


Title	Dynamic Safe Interruptibility for Decentralized Multi-Agent Reinforcement Learning
Authors	El Mahdi El Mhamdi, Rachid Guerraoui, Hadrien Hendrikx, Alexandre Maurer
Abstract	In reinforcement learning, agents learn by performing actions and observing their outcomes. Sometimes, it is desirable for a human operator to \textit{interrupt} an agent in order to prevent dangerous situations from happening. Yet, as part of their learning process, agents may link these interruptions, that impact their reward, to specific states and deliberately avoid them. The situation is particularly challenging in a multi-agent context because agents might not only learn from their own past interruptions, but also from those of other agents. Orseau and Armstrong defined \emph{safe interruptibility} for one learner, but their work does not naturally extend to multi-agent systems. This paper introduces \textit{dynamic safe interruptibility}, an alternative definition more suited to decentralized learning problems, and studies this notion in two learning frameworks: \textit{joint action learners} and \textit{independent learners}. We give realistic sufficient conditions on the learning algorithm to enable dynamic safe interruptibility in the case of joint action learners, yet show that these conditions are not sufficient for independent learners. We show however that if agents can detect interruptions, it is possible to prune the observations to ensure dynamic safe interruptibility even for independent learners.
Tasks	Multi-agent Reinforcement Learning
Published	2017-04-10
URL	http://arxiv.org/abs/1704.02882v2
PDF	http://arxiv.org/pdf/1704.02882v2.pdf
PWC	https://paperswithcode.com/paper/dynamic-safe-interruptibility-for
Repo
Framework

I2T2I: Learning Text to Image Synthesis with Textual Data Augmentation


Title	I2T2I: Learning Text to Image Synthesis with Textual Data Augmentation
Authors	Hao Dong, Jingqing Zhang, Douglas McIlwraith, Yike Guo
Abstract	Translating information between text and image is a fundamental problem in artificial intelligence that connects natural language processing and computer vision. In the past few years, performance in image caption generation has seen significant improvement through the adoption of recurrent neural networks (RNN). Meanwhile, text-to-image generation begun to generate plausible images using datasets of specific categories like birds and flowers. We’ve even seen image generation from multi-category datasets such as the Microsoft Common Objects in Context (MSCOCO) through the use of generative adversarial networks (GANs). Synthesizing objects with a complex shape, however, is still challenging. For example, animals and humans have many degrees of freedom, which means that they can take on many complex shapes. We propose a new training method called Image-Text-Image (I2T2I) which integrates text-to-image and image-to-text (image captioning) synthesis to improve the performance of text-to-image synthesis. We demonstrate that %the capability of our method to understand the sentence descriptions, so as to I2T2I can generate better multi-categories images using MSCOCO than the state-of-the-art. We also demonstrate that I2T2I can achieve transfer learning by using a pre-trained image captioning module to generate human images on the MPII Human Pose
Tasks	Data Augmentation, Image Captioning, Image Generation, Text-to-Image Generation, Transfer Learning
Published	2017-03-20
URL	http://arxiv.org/abs/1703.06676v3
PDF	http://arxiv.org/pdf/1703.06676v3.pdf
PWC	https://paperswithcode.com/paper/i2t2i-learning-text-to-image-synthesis-with
Repo
Framework

Solving a New 3D Bin Packing Problem with Deep Reinforcement Learning Method


Title	Solving a New 3D Bin Packing Problem with Deep Reinforcement Learning Method
Authors	Haoyuan Hu, Xiaodong Zhang, Xiaowei Yan, Longfei Wang, Yinghui Xu
Abstract	In this paper, a new type of 3D bin packing problem (BPP) is proposed, in which a number of cuboid-shaped items must be put into a bin one by one orthogonally. The objective is to find a way to place these items that can minimize the surface area of the bin. This problem is based on the fact that there is no fixed-sized bin in many real business scenarios and the cost of a bin is proportional to its surface area. Our research shows that this problem is NP-hard. Based on previous research on 3D BPP, the surface area is determined by the sequence, spatial locations and orientations of items. Among these factors, the sequence of items plays a key role in minimizing the surface area. Inspired by recent achievements of deep reinforcement learning (DRL) techniques, especially Pointer Network, on combinatorial optimization problems such as TSP, a DRL-based method is applied to optimize the sequence of items to be packed into the bin. Numerical results show that the method proposed in this paper achieve about 5% improvement than heuristic method.
Tasks	Combinatorial Optimization
Published	2017-08-20
URL	http://arxiv.org/abs/1708.05930v1
PDF	http://arxiv.org/pdf/1708.05930v1.pdf
PWC	https://paperswithcode.com/paper/solving-a-new-3d-bin-packing-problem-with
Repo
Framework

Maximum Entropy Distributions: Bit Complexity and Stability


Title	Maximum Entropy Distributions: Bit Complexity and Stability
Authors	Damian Straszak, Nisheeth K. Vishnoi
Abstract	Maximum entropy distributions with discrete support in $m$ dimensions arise in machine learning, statistics, information theory, and theoretical computer science. While structural and computational properties of max-entropy distributions have been extensively studied, basic questions such as: Do max-entropy distributions over a large support (e.g., $2^m$) with a specified marginal vector have succinct descriptions (polynomial-size in the input description)? and: Are entropy maximizing distributions “stable” under the perturbation of the marginal vector? have resisted a rigorous resolution. Here we show that these questions are related and resolve both of them. Our main result shows a ${\rm poly}(m, \log 1/\varepsilon)$ bound on the bit complexity of $\varepsilon$-optimal dual solutions to the maximum entropy convex program – for very general support sets and with no restriction on the marginal vector. Applications of this result include polynomial time algorithms to compute max-entropy distributions over several new and old polytopes for any marginal vector in a unified manner, a polynomial time algorithm to compute the Brascamp-Lieb constant in the rank-1 case. The proof of this result allows us to show that changing the marginal vector by $\delta$ changes the max-entropy distribution in the total variation distance roughly by a factor of ${\rm poly}(m, \log 1/\delta)\sqrt{\delta}$ – even when the size of the support set is exponential. Together, our results put max-entropy distributions on a mathematically sound footing – these distributions are robust and computationally feasible models for data.
Tasks
Published	2017-11-06
URL	https://arxiv.org/abs/1711.02036v2
PDF	https://arxiv.org/pdf/1711.02036v2.pdf
PWC	https://paperswithcode.com/paper/computing-maximum-entropy-distributions
Repo
Framework

All You Need is Beyond a Good Init: Exploring Better Solution for Training Extremely Deep Convolutional Neural Networks with Orthonormality and Modulation


Title	All You Need is Beyond a Good Init: Exploring Better Solution for Training Extremely Deep Convolutional Neural Networks with Orthonormality and Modulation
Authors	Di Xie, Jiang Xiong, Shiliang Pu
Abstract	Deep neural network is difficult to train and this predicament becomes worse as the depth increases. The essence of this problem exists in the magnitude of backpropagated errors that will result in gradient vanishing or exploding phenomenon. We show that a variant of regularizer which utilizes orthonormality among different filter banks can alleviate this problem. Moreover, we design a backward error modulation mechanism based on the quasi-isometry assumption between two consecutive parametric layers. Equipped with these two ingredients, we propose several novel optimization solutions that can be utilized for training a specific-structured (repetitively triple modules of Conv-BNReLU) extremely deep convolutional neural network (CNN) WITHOUT any shortcuts/ identity mappings from scratch. Experiments show that our proposed solutions can achieve distinct improvements for a 44-layer and a 110-layer plain networks on both the CIFAR-10 and ImageNet datasets. Moreover, we can successfully train plain CNNs to match the performance of the residual counterparts. Besides, we propose new principles for designing network structure from the insights evoked by orthonormality. Combined with residual structure, we achieve comparative performance on the ImageNet dataset.
Tasks
Published	2017-03-06
URL	http://arxiv.org/abs/1703.01827v3
PDF	http://arxiv.org/pdf/1703.01827v3.pdf
PWC	https://paperswithcode.com/paper/all-you-need-is-beyond-a-good-init-exploring
Repo
Framework

Synth-Validation: Selecting the Best Causal Inference Method for a Given Dataset


Title	Synth-Validation: Selecting the Best Causal Inference Method for a Given Dataset
Authors	Alejandro Schuler, Ken Jung, Robert Tibshirani, Trevor Hastie, Nigam Shah
Abstract	Many decisions in healthcare, business, and other policy domains are made without the support of rigorous evidence due to the cost and complexity of performing randomized experiments. Using observational data to answer causal questions is risky: subjects who receive different treatments also differ in other ways that affect outcomes. Many causal inference methods have been developed to mitigate these biases. However, there is no way to know which method might produce the best estimate of a treatment effect in a given study. In analogy to cross-validation, which estimates the prediction error of predictive models applied to a given dataset, we propose synth-validation, a procedure that estimates the estimation error of causal inference methods applied to a given dataset. In synth-validation, we use the observed data to estimate generative distributions with known treatment effects. We apply each causal inference method to datasets sampled from these distributions and compare the effect estimates with the known effects to estimate error. Using simulations, we show that using synth-validation to select a causal inference method for each study lowers the expected estimation error relative to consistently using any single method.
Tasks	Causal Inference
Published	2017-10-31
URL	http://arxiv.org/abs/1711.00083v1
PDF	http://arxiv.org/pdf/1711.00083v1.pdf
PWC	https://paperswithcode.com/paper/synth-validation-selecting-the-best-causal
Repo
Framework

Entropy-guided Retinex anisotropic diffusion algorithm based on partial differential equations (PDE) for illumination correction


Title	Entropy-guided Retinex anisotropic diffusion algorithm based on partial differential equations (PDE) for illumination correction
Authors	U. A. Nnolim
Abstract	This report describes the experimental results obtained using a proposed variational Retinex algorithm for controlled illumination correction. Two colour restoration and enhancement schemes of the algorithm are presented for drastically improved results. The algorithm modifies the reflectance image using global and local contrast enhancement approaches and gradually removes the residual illumination to yield highly pleasing results. The proposed algorithms are optimized by way of simultaneous perceptual quality metric (PQM) stabilization and entropy maximization for fully automated processing solving the problem of determination of stopping time. The usage of the HSI or HSV colour space ensures a unique solution to the optimization problem unlike in the RGB space where there is none (forcing manual selection of number of iteration. The proposed approach preserves and enhances details in both bright and dark regions of underexposed images in addition to eliminating the colour distortion, over-exposure in bright image regions, halo effect and grey-world violations observed in Retinex-based approaches. Extensive experiments indicate consistent performance as the proposed approach exploits and augments the advantages of PDE-based formulation, performing illumination correction, colour enhancement correction and restoration, contrast enhancement and noise suppression. Comparisons shows that the proposed approach surpasses most of the other conventional algorithms found in the literature.
Tasks
Published	2017-02-04
URL	http://arxiv.org/abs/1702.01339v1
PDF	http://arxiv.org/pdf/1702.01339v1.pdf
PWC	https://paperswithcode.com/paper/entropy-guided-retinex-anisotropic-diffusion
Repo
Framework

SafetyNets: Verifiable Execution of Deep Neural Networks on an Untrusted Cloud


Title	SafetyNets: Verifiable Execution of Deep Neural Networks on an Untrusted Cloud
Authors	Zahra Ghodsi, Tianyu Gu, Siddharth Garg
Abstract	Inference using deep neural networks is often outsourced to the cloud since it is a computationally demanding task. However, this raises a fundamental issue of trust. How can a client be sure that the cloud has performed inference correctly? A lazy cloud provider might use a simpler but less accurate model to reduce its own computational load, or worse, maliciously modify the inference results sent to the client. We propose SafetyNets, a framework that enables an untrusted server (the cloud) to provide a client with a short mathematical proof of the correctness of inference tasks that they perform on behalf of the client. Specifically, SafetyNets develops and implements a specialized interactive proof (IP) protocol for verifiable execution of a class of deep neural networks, i.e., those that can be represented as arithmetic circuits. Our empirical results on three- and four-layer deep neural networks demonstrate the run-time costs of SafetyNets for both the client and server are low. SafetyNets detects any incorrect computations of the neural network by the untrusted server with high probability, while achieving state-of-the-art accuracy on the MNIST digit recognition (99.4%) and TIMIT speech recognition tasks (75.22%).
Tasks	Speech Recognition
Published	2017-06-30
URL	http://arxiv.org/abs/1706.10268v1
PDF	http://arxiv.org/pdf/1706.10268v1.pdf
PWC	https://paperswithcode.com/paper/safetynets-verifiable-execution-of-deep
Repo
Framework

Unmixing dynamic PET images with variable specific binding kinetics


Title	Unmixing dynamic PET images with variable specific binding kinetics
Authors	Yanna Cruz Cavalcanti, Thomas Oberlin, Nicolas Dobigeon, Simon Stute, Maria Ribeiro, Clovis Tauber
Abstract	To analyze dynamic positron emission tomography (PET) images, various generic multivariate data analysis techniques have been considered in the literature, such as principal component analysis (PCA), independent component analysis (ICA), factor analysis and nonnegative matrix factorization (NMF). Nevertheless, these conventional approaches neglect any possible nonlinear variations in the time activity curves describing the kinetic behavior of tissues with specific binding, which limits their ability to recover a reliable, understandable and interpretable description of the data. This paper proposes an alternative analysis paradigm that accounts for spatial fluctuations in the exchange rate of the tracer between a free compartment and a specifically bound ligand compartment. The method relies on the concept of linear unmixing, usually applied on the hyperspectral domain, which combines NMF with a sum-to-one constraint that ensures an exhaustive description of the mixtures. The spatial variability of the signature corresponding to the specific binding tissue is explicitly modeled through a perturbed component. The performance of the method is assessed on both synthetic and real data and is shown to compete favorably when compared to other conventional analysis methods. The proposed method improved both factor estimation and proportions extraction for specific binding. Modeling the variability of the specific binding factor has a strong potential impact for dynamic PET image analysis.
Tasks
Published	2017-07-19
URL	http://arxiv.org/abs/1707.09867v2
PDF	http://arxiv.org/pdf/1707.09867v2.pdf
PWC	https://paperswithcode.com/paper/unmixing-dynamic-pet-images-with-variable
Repo
Framework

Multitask Learning with Low-Level Auxiliary Tasks for Encoder-Decoder Based Speech Recognition


Title	Multitask Learning with Low-Level Auxiliary Tasks for Encoder-Decoder Based Speech Recognition
Authors	Shubham Toshniwal, Hao Tang, Liang Lu, Karen Livescu
Abstract	End-to-end training of deep learning-based models allows for implicit learning of intermediate representations based on the final task loss. However, the end-to-end approach ignores the useful domain knowledge encoded in explicit intermediate-level supervision. We hypothesize that using intermediate representations as auxiliary supervision at lower levels of deep networks may be a good way of combining the advantages of end-to-end training and more traditional pipeline approaches. We present experiments on conversational speech recognition where we use lower-level tasks, such as phoneme recognition, in a multitask training approach with an encoder-decoder model for direct character transcription. We compare multiple types of lower-level tasks and analyze the effects of the auxiliary tasks. Our results on the Switchboard corpus show that this approach improves recognition accuracy over a standard encoder-decoder model on the Eval2000 test set.
Tasks	Speech Recognition
Published	2017-04-05
URL	http://arxiv.org/abs/1704.01631v2
PDF	http://arxiv.org/pdf/1704.01631v2.pdf
PWC	https://paperswithcode.com/paper/multitask-learning-with-low-level-auxiliary
Repo
Framework