ICLR 2019 Papers

Layout:

mini compact topic detail

Exemplar Guided Unsupervised Image-to-Image Translation with Semantic Consistency

Variance Reduction for Reinforcement Learning in Input-Driven Environments

Improving Differentiable Neural Computers Through Memory Masking, De-allocation, and Link Distribution Sharpness Control

Learning from Positive and Unlabeled Data with a Selection Bias

Representing Formal Languages: A Comparison Between Finite Automata and Recurrent Neural Networks

INVASE: Instance-wise Variable Selection using Neural Networks

Two-Timescale Networks for Nonlinear Value Function Approximation

Deterministic Variational Inference for Robust Bayesian Neural Networks

Analyzing Inverse Problems with Invertible Neural Networks

Integer Networks for Data Compression with Latent-Variable Models

Minimal Random Code Learning: Getting Bits Back from Compressed Model Parameters

Do Deep Generative Models Know What They Don't Know?

Learning to Screen for Fast Softmax Inference on Large Vocabulary Neural Networks

Dimensionality Reduction for Representing the Knowledge of Probabilistic Models

Deep Layers as Stochastic Solvers

KnockoffGAN: Generating Knockoffs for Feature Selection using Generative Adversarial Networks

Multiple-Attribute Text Rewriting

Smoothing the Geometry of Probabilistic Box Embeddings

Critical Learning Periods in Deep Networks

RNNs implicitly implement tensor-product representations

InfoBot: Transfer and Exploration via the Information Bottleneck

Adversarial Imitation via Variational Inverse Reinforcement Learning

Rigorous Agent Evaluation: An Adversarial Approach to Uncover Catastrophic Failures

Maximal Divergence Sequential Autoencoder for Binary Software Vulnerability Detection

Beyond Pixel Norm-Balls: Parametric Adversaries using an Analytically Differentiable Renderer

Towards the first adversarially robust neural network model on MNIST

Diffusion Scattering Transforms on Graphs

A Variational Inequality Perspective on Generative Adversarial Networks

Cost-Sensitive Robustness against Adversarial Examples

STCN: Stochastic Temporal Convolutional Networks

Adversarial Reprogramming of Neural Networks

Towards GAN Benchmarks Which Require Generalization

Feed-forward Propagation in Probabilistic Neural Networks with Categorical and Max Layers

Model-Predictive Policy Learning with Uncertainty Regularization for Driving in Dense Traffic

Three Mechanisms of Weight Decay Regularization

Efficient Augmentation via Data Subsampling

Differentiable Learning-to-Normalize via Switchable Normalization

Context-adaptive Entropy Model for End-to-end Optimized Image Compression

Optimal Transport Maps For Distribution Preserving Operations on Latent Spaces of Generative Models

Characterizing Audio Adversarial Examples Using Temporal Dependency

Neural TTS Stylization with Adversarial and Collaborative Games

Overcoming the Disentanglement vs Reconstruction Trade-off via Jacobian Supervision

Hierarchical Reinforcement Learning via Advantage-Weighted Information Maximization

Human-level Protein Localization with Convolutional Neural Networks

Learning Programmatically Structured Representations with Perceptor Gradients

How to train your MAML

Robust Conditional Generative Adversarial Networks

Explaining Image Classifiers by Counterfactual Generation

Double Viterbi: Weight Encoding for High Compression Ratio and Fast On-Chip Reconstruction for Deep Neural Network

Von Mises-Fisher Loss for Training Sequence to Sequence Models with Continuous Outputs

Learning protein sequence embeddings using information from structure

Bounce and Learn: Modeling Scene Dynamics with Real-World Bounces

StrokeNet: A Neural Painting Environment

Discovery of Natural Language Concepts in Individual Units of CNNs

MisGAN: Learning from Incomplete Data with Generative Adversarial Networks

Solving the Rubik's Cube with Approximate Policy Iteration

Exploration by random network distillation

Preferences Implicit in the State of the World

Competitive experience replay

Efficient Multi-Objective Neural Architecture Search via Lamarckian Evolution

GAN Dissection: Visualizing and Understanding Generative Adversarial Networks

ProbGAN: Towards Probabilistic GAN with Theoretical Guarantees

No Training Required: Exploring Random Encoders for Sentence Classification

Don't let your Discriminator be fooled

Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interference

Temporal Difference Variational Auto-Encoder

A Generative Model For Electron Paths

AD-VAT: An Asymmetric Dueling mechanism for learning Visual Active Tracking

Sliced Wasserstein Auto-Encoders

MARGINALIZED AVERAGE ATTENTIONAL NETWORK FOR WEAKLY-SUPERVISED LEARNING

Value Propagation Networks

Woulda, Coulda, Shoulda: Counterfactually-Guided Policy Search

Learning to Adapt in Dynamic, Real-World Environments through Meta-Reinforcement Learning

Directed-Info GAIL: Learning Hierarchical Policies from Unsegmented Demonstrations using Directed Information

Meta-Learning Probabilistic Inference for Prediction

Learning to Represent Edits

textTOvec: DEEP CONTEXTUALIZED NEURAL AUTOREGRESSIVE TOPIC MODELS OF LANGUAGE WITH DISTRIBUTED COMPOSITIONAL PRIOR

An Empirical study of Binary Neural Networks' Optimisation

Building Dynamic Knowledge Graphs from Text using Machine Reading Comprehension

L-Shapley and C-Shapley: Efficient Model Interpretation for Structured Data

GENERATING HIGH FIDELITY IMAGES WITH SUBSCALE PIXEL NETWORKS AND MULTIDIMENSIONAL UPSCALING

Learning Finite State Representations of Recurrent Policy Networks

Variational Bayesian Phylogenetic Inference

Supervised Community Detection with Line Graph Neural Networks

Interpolation-Prediction Networks for Irregularly Sampled Time Series

Information Theoretic lower bounds on negative log likelihood

Measuring Compositionality in Representation Learning

CBOW Is Not All You Need: Combining CBOW with the Compositional Matrix Space Model

A Data-Driven and Distributed Approach to Sparse Signal Representation and Recovery

Attention, Learn to Solve Routing Problems!

Wasserstein Barycenter Model Ensembling

Preventing Posterior Collapse with delta-VAEs

Information asymmetry in KL-regularized RL

Multi-Agent Dual Learning

Neural Speed Reading with Structural-Jump-LSTM

Pay Less Attention with Lightweight and Dynamic Convolutions

Guiding Policies with Language via Meta-Learning

From Language to Goals: Inverse Reinforcement Learning for Vision-Based Instruction Following

Large-Scale Answerer in Questioner's Mind for Visual Dialog Question Generation

Learning to Learn with Conditional Class Dependencies

M^3RL: Mind-aware Multi-agent Management Reinforcement Learning

LEARNING FACTORIZED REPRESENTATIONS FOR OPEN-SET DOMAIN ADAPTATION

FUNCTIONAL VARIATIONAL BAYESIAN NEURAL NETWORKS

A new dog learns old tricks: RL finds classic optimization algorithms

Recall Traces: Backtracking Models for Efficient Reinforcement Learning

Robustness May Be at Odds with Accuracy

CEM-RL: Combining evolutionary and gradient-based methods for policy search

ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware

What do you learn from context? Probing for sentence structure in contextualized word representations

Music Transformer: Generating Music with Long-Term Structure

Learning what and where to attend

Attentive Neural Processes

Spreading vectors for similarity search

Diversity and Depth in Per-Example Routing Models

Multi-step Retriever-Reader Interaction for Scalable Open-domain Question Answering

Generative Code Modeling with Graphs

Unsupervised Hyper-alignment for Multilingual Word Embeddings

Structured Neural Summarization

Eidetic 3D LSTM: A Model for Video Prediction and Beyond

Auxiliary Variational MCMC

Deep reinforcement learning with relational inductive biases

TimbreTron: A WaveNet(CycleGAN(CQT(Audio))) Pipeline for Musical Timbre Transfer

A Universal Music Translation Network

Whitening and Coloring Batch Transform for GANs

From Hard to Soft: Understanding Deep Network Nonlinearities via Vector Quantization and Statistical Inference

Aggregated Momentum: Stability Through Passive Damping

Deep learning generalizes because the parameter-function map is biased towards simple functions

Adaptive Posterior Learning: few-shot learning with a surprise-based memory module

Generating Multi-Agent Trajectories using Programmatic Weak Supervision

Differentiable Perturb-and-Parse: Semi-Supervised Parsing with a Structured Variational Autoencoder

Information-Directed Exploration for Deep Reinforcement Learning

The Singular Values of Convolutional Layers

On Self Modulation for Generative Adversarial Networks

A Closer Look at Few-shot Classification

Local SGD Converges Fast and Communicates Little

Adaptive Estimators Show Information Compression in Deep Neural Networks

LEARNING TO PROPAGATE LABELS: TRANSDUCTIVE PROPAGATION NETWORK FOR FEW-SHOT LEARNING

Deep Frank-Wolfe For Neural Network Optimization

Adaptive Gradient Methods with Dynamic Bound of Learning Rate

Deep Decoder: Concise Image Representations from Untrained Non-convolutional Networks

The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks

Relational Forward Models for Multi-Agent Learning

ANYTIME MINIBATCH: EXPLOITING STRAGGLERS IN ONLINE DISTRIBUTED OPTIMIZATION

Analysis of Quantized Models

Initialized Equilibrium Propagation for Backprop-Free Training

RelGAN: Relational Generative Adversarial Networks for Text Generation

Energy-Constrained Compression for Deep Neural Networks via Weighted Sparse Projection and Layer Input Masking

Towards Robust, Locally Linear Deep Networks

Bias-Reduced Uncertainty Estimation for Deep Neural Classifiers

Spectral Inference Networks: Unifying Deep and Spectral Learning

Toward Understanding the Impact of Staleness in Distributed Machine Learning

Theoretical Analysis of Auto Rate-Tuning by Batch Normalization

Learning Protein Structure with a Differentiable Simulator

Generalizable Adversarial Training via Spectral Normalization

Efficiently testing local optimality and escaping saddles for ReLU networks

Gradient Descent Provably Optimizes Over-parameterized Neural Networks

Universal Stagewise Learning for Non-Convex Problems with Convergence on Averaged Solutions

Deterministic PAC-Bayesian generalization bounds for deep networks via generalizing noise-resilience

Variational Autoencoders with Jointly Optimized Latent Dependency Structure

Multi-Domain Adversarial Learning

Gradient descent aligns the layers of deep linear networks

Adaptive Input Representations for Neural Language Modeling

On Random Deep Weight-Tied Autoencoders: Exact Asymptotic Analysis, Phase Transitions, and Implications to Training

Learning Neural PDE Solvers with Convergence Guarantees

Marginal Policy Gradients: A Unified Family of Estimators for Bounded Action Spaces with Applications

Deep Learning 3D Shapes Using Alt-az Anisotropic 2-Sphere Convolution

DeepOBS: A Deep Learning Optimizer Benchmark Suite

BA-Net: Dense Bundle Adjustment Networks

ARM: Augment-REINFORCE-Merge Gradient for Stochastic Binary Networks

Equi-normalization of Neural Networks

Amortized Bayesian Meta-Learning

Harmonizing Maximum Likelihood with GANs for Multimodal Conditional Generation

Decoupled Weight Decay Regularization

Analysing Mathematical Reasoning Abilities of Neural Models

Learning to Schedule Communication in Multi-agent Reinforcement Learning

Large-Scale Study of Curiosity-Driven Learning

Knowledge Flow: Improve Upon Your Teachers

Emergent Coordination Through Competition

How Important is a Neuron

Backpropamine: training self-modifying neural networks with differentiable neuromodulated plasticity

Learning Factorized Multimodal Representations

Learning To Simulate

Janossy Pooling: Learning Deep Permutation-Invariant Functions for Variable-Size Inputs

A rotation-equivariant convolutional neural network model of primary visual cortex

Learning to Design RNA

Stochastic Optimization of Sorting Networks via Continuous Relaxations

Neural Graph Evolution: Automatic Robot Design

Policy Transfer with Strategy Optimization

Selfless Sequential Learning

Overcoming Catastrophic Forgetting for Continual Learning via Model Adaptation

Learning Multi-Level Hierarchies with Hindsight

Generating Liquid Simulations with Deformation-aware Neural Networks

Neural Persistence: A Complexity Measure for Deep Neural Networks Using Algebraic Topology

A Direct Approach to Robust Deep Learning Using Adversarial Networks

Plan Online, Learn Offline: Efficient Learning and Exploration via Model-Based Control

Unsupervised Adversarial Image Reconstruction

Deep, Skinny Neural Networks are not Universal Approximators

Subgradient Descent Learns Orthogonal Dictionaries

There Are Many Consistent Explanations of Unlabeled Data: Why You Should Average

SNIP: SINGLE-SHOT NETWORK PRUNING BASED ON CONNECTION SENSITIVITY

Unsupervised Learning via Meta-Learning

Excessive Invariance Causes Adversarial Vulnerability

Diagnosing and Enhancing VAE Models

K for the Price of 1: Parameter-efficient Multi-task and Transfer Learning

Improving Generalization and Stability of Generative Adversarial Networks

Adaptivity of deep ReLU network for learning in Besov and mixed smooth Besov spaces: optimal rate and curse of dimensionality

ACCELERATING NONCONVEX LEARNING VIA REPLICA EXCHANGE LANGEVIN DIFFUSION

Stochastic Prediction of Multi-Agent Interactions from Partial Observations

Adversarial Audio Synthesis

Hindsight policy gradients

Efficient Training on Very Large Corpora via Gramian Estimation

The Deep Weight Prior

Tree-Structured Recurrent Switching Linear Dynamical Systems for Multi-Scale Modeling

Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees

The role of over-parametrization in generalization of neural networks

Are adversarial examples inevitable?

Variance Networks: When Expectation Does Not Meet Your Expectations

SGD Converges to Global Minimum in Deep Learning via Star-convex Path

Learning concise representations for regression by evolving networks of trees

PATE-GAN: Generating Synthetic Data with Differential Privacy Guarantees

Verification of Non-Linear Specifications for Neural Networks

Rethinking the Value of Network Pruning

signSGD with Majority Vote is Communication Efficient and Fault Tolerant

On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length

Learning-Based Frequency Estimation Algorithms

Learning Two-layer Neural Networks with Symmetric Inputs

signSGD via Zeroth-Order Oracle

RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space

Neural Program Repair by Jointly Learning to Localize and Repair

Regularized Learning for Domain Adaptation under Label Shifts

On Computation and Generalization of Generative Adversarial Networks under Spectrum Control

Distribution-Interpolation Trade off in Generative Models

Automatically Composing Representation Transformations as a Means for Generalization

Defensive Quantization: When Efficiency Meets Robustness

Detecting Egregious Responses in Neural Sequence-to-sequence Models

Combinatorial Attacks on Binarized Neural Networks

Structured Adversarial Attack: Towards General Implementation and Better Interpretability

Posterior Attention Models for Sequence to Sequence Learning

Dynamic Channel Pruning: Feature Boosting and Suppression

Big-Little Net: An Efficient Multi-Scale Feature Representation for Visual and Speech Recognition

Relaxed Quantization for Discretized Neural Networks

PeerNets: Exploiting Peer Wisdom Against Adversarial Attacks

Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluids

Prior Convictions: Black-box Adversarial Attacks with Bandits and Priors

Neural network gradient-based learning of black-box function interfaces

Learning Mixed-Curvature Representations in Product Spaces

NOODL: Provable Online Dictionary Learning and Sparse Coding

Sparse Dictionary Learning by Dynamical Neural Networks

GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding

The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervision

DISTRIBUTIONAL CONCAVITY REGULARIZATION FOR GANS

Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile

Meta-Learning with Latent Embedding Optimization

Learning Latent Superstructures in Variational Autoencoders for Deep Multidimensional Clustering

Measuring and regularizing networks in function space

Towards Understanding Regularization in Batch Normalization

Learnable Embedding Space for Efficient Neural Architecture Compression

Mode Normalization

Synthetic Datasets for Neural Program Synthesis

Improving Sequence-to-Sequence Learning via Optimal Transport

A Max-Affine Spline Perspective of Recurrent Neural Networks

Unsupervised Control Through Non-Parametric Discriminative Rewards

ProxQuant: Quantized Neural Networks via Proximal Operators

Beyond Greedy Ranking: Slate Optimization via List-CVAE

Adversarial Attacks on Graph Neural Networks via Meta Learning

Trellis Networks for Sequence Modeling

Self-Tuning Networks: Bilevel Optimization of Hyperparameters using Structured Best-Response Functions

Bayesian Policy Optimization for Model Uncertainty

Soft Q-Learning with Mutual-Information Regularization

Self-Monitoring Navigation Agent via Auxiliary Progress Estimation

Visual Semantic Navigation using Scene Priors

Minimum Divergence vs. Maximum Margin: an Empirical Comparison on Seq2Seq Models

Boosting Robustness Certification of Neural Networks

Neural Logic Machines

Universal Successor Features Approximators

Generative predecessor models for sample-efficient imitation learning

Learning Embeddings into Entropic Wasserstein Spaces

Conditional Network Embeddings

Imposing Category Trees Onto Word-Embeddings Using A Geometric Construction

Meta-Learning Update Rules for Unsupervised Representation Learning

Small nonlinearities in activation functions create bad local minima in neural networks

On the Minimal Supervision for Training Any Binary Classifier from Only Unlabeled Data

Non-vacuous Generalization Bounds at the ImageNet Scale: a PAC-Bayesian Compression Approach

Hyperbolic Attention Networks

The Comparative Power of ReLU Networks and Polynomial Kernels in the Presence of Sparse Latent Structure

ALISTA: Analytic Weights Are As Good As Learned Weights in LISTA

Learning a Meta-Solver for Syntax-Guided Program Synthesis

DyRep: Learning Representations over Dynamic Graphs

Understanding and Improving Interpolation in Autoencoders via an Adversarial Regularizer

An analytic theory of generalization dynamics and transfer learning in deep linear networks

Multilingual Neural Machine Translation with Knowledge Distillation

Slimmable Neural Networks

Residual Non-local Attention Networks for Image Restoration

Learning Robust Representations by Projecting Superficial Statistics Out

Composing Complex Skills by Learning Transition Policies

How Powerful are Graph Neural Networks?

Episodic Curiosity through Reachability

A Convergence Analysis of Gradient Descent for Deep Linear Neural Networks

Poincare Glove: Hyperbolic Word Embeddings

Sample Efficient Adaptive Text-to-Speech

A Kernel Random Matrix-Based Approach for Sparse PCA

Fluctuation-dissipation relations for stochastic gradient descent

Probabilistic Planning with Sequential Monte Carlo methods

Deep Graph Infomax

GamePad: A Learning Environment for Theorem Proving

Augmented Cyclic Adversarial Learning for Low Resource Domain Adaptation

SPIGAN: Privileged Adversarial Learning from Simulation

FFJORD: Free-Form Continuous Dynamics for Scalable Reversible Generative Models

Systematic Generalization: What Is Required and Can It Be Learned?

Modeling Uncertainty with Hedged Instance Embeddings

DHER: Hindsight Experience Replay for Dynamic Goals

On the Convergence of A Class of Adam-Type Algorithms for Non-Convex Optimization

Latent Convolutional Models

Deep Anomaly Detection with Outlier Exposure

Transferring Knowledge across Learning Processes

Learning when to Communicate at Scale in Multiagent Cooperative and Competitive Tasks

Learning to Describe Scenes with Programs

Unsupervised Discovery of Parts, Structure, and Dynamics

A Statistical Approach to Assessing Neural Network Robustness

On the loss landscape of a class of deep neural networks with no bad local valleys

Harmonic Unpaired Image-to-image Translation

Invariant and Equivariant Graph Networks

LeMoNADe: Learned Motif and Neuronal Assembly Detection in calcium imaging videos

Preconditioner on Matrix Lie Group for SGD

FlowQA: Grasping Flow in History for Conversational Machine Comprehension

Execution-Guided Neural Program Synthesis

Representation Degeneration Problem in Training Natural Language Generation Models

Opportunistic Learning: Budgeted Cost-Sensitive Learning from Data Streams

Coarse-grain Fine-grain Coattention Network for Multi-evidence Question Answering

Reward Constrained Policy Optimization

Doubly Reparameterized Gradient Estimators for Monte Carlo Objectives

Hierarchical Generative Modeling for Controllable Speech Synthesis

An Empirical Study of Example Forgetting during Deep Neural Network Learning

Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow

Deep Convolutional Networks as shallow Gaussian Processes

Learning To Solve Circuit-SAT: An Unsupervised Differentiable Approach

Towards Metamerism via Foveated Style Transfer

Generalized Tensor Models for Recurrent Neural Networks

Per-Tensor Fixed-Point Quantization of the Back-Propagation Algorithm

Understanding Composition of Word Embeddings via Tensor Decomposition

Unsupervised Learning of the Set of Local Maxima

Supervised Policy Update for Deep Reinforcement Learning

DialogWAE: Multimodal Response Generation with Conditional Wasserstein Auto-Encoder

InstaGAN: Instance-aware Image-to-Image Translation

GO Gradient for Expectation-Based Objectives

Graph HyperNetworks for Neural Architecture Search

Wizard of Wikipedia: Knowledge-Powered Conversational Agents

Optimal Completion Distillation for Sequence Learning

Multi-class classification without multi-class labels

Generating Multiple Objects at Spatially Distinct Locations

Transfer Learning for Sequences via Learning to Collocate

Learning to Understand Goal Specifications by Modelling Reward

BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learning

A Unified Theory of Early Visual Representations from Retina to Cortex through Anatomically Constrained Deep CNNs

Benchmarking Neural Network Robustness to Common Corruptions and Perturbations

The Limitations of Adversarial Training and the Blind-Spot Attack

SOM-VAE: Interpretable Discrete Representation Learning on Time Series

The relativistic discriminator: a key element missing from standard GAN

Reasoning About Physical Interactions with Object-Oriented Prediction and Planning

Learning a SAT Solver from Single-Bit Supervision

MAE: Mutual Posterior-Divergence Regularization for Variational AutoEncoders

G-SGD: Optimizing ReLU Neural Networks in its Positively Scale-Invariant Space

NADPEx: An on-policy temporally consistent exploration method for deep reinforcement learning

Optimal Control Via Neural Networks: A Convex Approach

Diversity is All You Need: Learning Skills without a Reward Function

Learning Multimodal Graph-to-Graph Translation for Molecule Optimization

AdaShift: Decorrelation and Convergence of Adaptive Learning Rate Methods

Post Selection Inference with Incomplete Maximum Mean Discrepancy Estimator

Quasi-hyperbolic momentum and Adam for deep learning

Efficient Lifelong Learning with A-GEM

Quaternion Recurrent Neural Networks

A Closer Look at Deep Learning Heuristics: Learning rate restarts, Warmup and Distillation

Graph Wavelet Neural Network

Convolutional Neural Networks on Non-uniform Geometrical Signals Using Euclidean Spectral Transformation

Spherical CNNs on Unstructured Grids

On the Turing Completeness of Modern Neural Network Architectures

ADef: an Iterative Algorithm to Construct Adversarial Deformations

Don't Settle for Average, Go for the Max: Fuzzy Sets and Max-Pooled Word Vectors

Random mesh projectors for inverse problems

Bayesian Deep Convolutional Networks with Many Channels are Gaussian Processes

Learning Grid Cells as Vector Representation of Self-Position Coupled with Matrix Representation of Self-Motion

SNAS: stochastic neural architecture search

On the Universal Approximability and Complexity Bounds of Quantized ReLU Neural Networks

Feature Intertwiner for Object Detection

ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness

AntisymmetricRNN: A Dynamical System View on Recurrent Neural Networks

Contingency-Aware Exploration in Reinforcement Learning

Unsupervised Domain Adaptation for Distance Metric Learning

Kernel Change-point Detection with Auxiliary Deep Generative Models

Learning Representations of Sets through Optimized Permutations

Global-to-local Memory Pointer Networks for Task-Oriented Dialogue

Practical lossless compression with latent variables using bits back coding

Stochastic Gradient/Mirror Descent: Minimax Optimality and Implicit Regularization

Emerging Disentanglement in Auto-Encoder Based Unsupervised Image Content Transfer

Adversarial Domain Adaptation for Stable Brain-Machine Interfaces

Hierarchical Visuomotor Control of Humanoids

Neural Probabilistic Motor Primitives for Humanoid Control

h-detach: Modifying the LSTM Gradient Towards Better Optimization

DPSNet: End-to-end Deep Plane Sweep Stereo

Hierarchical interpretations for neural network predictions

ClariNet: Parallel Wave Generation in End-to-End Text-to-Speech

Minimal Images in Deep Neural Networks: Fragile Object Recognition in Natural Images

Complement Objective Training

Variational Smoothing in Recurrent Neural Network Language Models

Improving MMD-GAN Training with Repulsive Loss Function

Improving the Generalization of Adversarial Training with Domain Adaptation

Label super-resolution networks

Enabling Factorized Piano Music Modeling and Generation with the MAESTRO Dataset

Understanding Straight-Through Estimator in Training Activation Quantized Neural Nets

Bayesian Prediction of Future Street Scenes using Synthetic Likelihoods

Meta-learning with differentiable closed-form solvers

Approximating CNNs with Bag-of-local-Features models works surprisingly well on ImageNet

Learning to Remember More with Less Memorization

Max-MIG: an Information Theoretic Approach for Joint Learning from Crowds

Accumulation Bit-Width Scaling For Ultra-Low Precision Training Of Deep Networks

Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Network

Learning Actionable Representations with Goal Conditioned Policies

Biologically-Plausible Learning Algorithms Can Scale to Large Datasets

Large Scale Graph Learning From Smooth Signals

Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks

Variational Autoencoder with Arbitrary Conditioning

Stable Recurrent Models

A comprehensive, application-oriented study of catastrophic forgetting in DNNs

Identifying and Controlling Important Neurons in Neural Machine Translation

Probabilistic Recursive Reasoning for Multi-Agent Reinforcement Learning

Learning Recurrent Binary/Ternary Weights

Learning what you can do before doing anything

Dynamic Sparse Graph for Efficient Deep Learning

Learning deep representations by mutual information estimation and maximization

Recurrent Experience Replay in Distributed Reinforcement Learning

Visual Reasoning by Progressive Module Networks

code2seq: Generating Sequences from Structured Representations of Code

CAMOU: Learning Physical Vehicle Camouflages to Adversarially Attack Detectors in the Wild

DELTA: DEEP LEARNING TRANSFER USING FEATURE MAP WITH ATTENTION FOR CONVOLUTIONAL NETWORKS

Fixup Initialization: Residual Learning Without Normalization

Stable Opponent Shaping in Differentiable Games

Unsupervised Speech Recognition via Segmental Empirical Output Distribution Matching

Lagging Inference Networks and Posterior Collapse in Variational Autoencoders

Learning to Infer and Execute 3D Shape Programs

Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware

GANSynth: Adversarial Neural Audio Synthesis

Learning Self-Imitating Diverse Policies

Riemannian Adaptive Optimization Methods

AutoLoss: Learning Discrete Schedule for Alternate Optimization

Off-Policy Evaluation and Learning from Logged Bandit Feedback: Error Reduction via Surrogate Policy

Function Space Particle Optimization for Bayesian Neural Networks

LanczosNet: Multi-Scale Deep Graph Convolutional Networks

Meta-Learning For Stochastic Gradient MCMC

Training for Faster Adversarial Robustness Verification via Inducing ReLU Stability

Capsule Graph Neural Network

Time-Agnostic Prediction: Predicting Predictable Video Frames

Learning sparse relational transition models

Large Scale GAN Training for High Fidelity Natural Image Synthesis

Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds

RotDCF: Decomposition of Convolutional Filters for Rotation-Equivariant Deep Networks

DOM-Q-NET: Grounded RL on Structured Language

Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learning

The Unusual Effectiveness of Averaging in GAN Training

Deep Online Learning Via Meta-Learning: Continual Adaptation for Model-Based RL

On the Sensitivity of Adversarial Robustness to Input Data Distributions

Generative Question Answering: Learning to Answer the Whole Question

Query-Efficient Hard-label Black-box Attack: An Optimization-based Approach

ProMP: Proximal Meta-Policy Search

Near-Optimal Representation Learning for Hierarchical Reinforcement Learning

Revealing interpretable object representations from human behavior

LayoutGAN: Generating Graphic Layouts with Wireframe Discriminators

Caveats for information bottleneck in deterministic scenarios

Modeling the Long Term Future in Model-Based Reinforcement Learning

Sample Efficient Imitation Learning for Continuous Control

Dynamically Unfolding Recurrent Restorer: A Moving Endpoint Control Method for Image Restoration

Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learning

Learning Procedural Abstractions and Evaluating Discrete Latent Temporal Structure

Disjoint Mapping Network for Cross-modal Matching of Voices and Faces

Diversity-Sensitive Conditional Generative Adversarial Networks

Feature-Wise Bias Amplification

L2-Nonexpansive Neural Networks

Kernel RNN Learning (KeRNL)

Multilingual Neural Machine Translation With Soft Decoupled Encoding

Hierarchical RL Using an Ensemble of Proprioceptive Periodic Policies

Learning to Navigate the Web

Top-Down Neural Model For Formulae

A2BCD: Asynchronous Acceleration with Optimal Complexity

The Laplacian in RL: Learning Representations with Efficient Approximations

A Mean Field Theory of Batch Normalization

DARTS: Differentiable Architecture Search

Learning to Make Analogies by Contrasting Abstract Relational Structure

Discriminator Rejection Sampling

Approximability of Discriminators Implies Diversity in GANs

Environment Probing Interaction Policies

Predicting the Generalization Gap in Deep Networks with Margin Distributions

Learning Exploration Policies for Navigation

Visual Explanation by Interpretation: Improving Visual Feedback Capabilities of Deep Neural Networks

Scalable Unbalanced Optimal Transport using Generative Adversarial Networks

Active Learning with Partial Feedback

ROBUST ESTIMATION VIA GENERATIVE ADVERSARIAL NETWORKS

Evaluating Robustness of Neural Networks with Mixed Integer Programming

Visceral Machines: Risk-Aversion in Reinforcement Learning with Intrinsic Physiological Rewards

Learning Localized Generative Models for 3D Point Clouds via Graph Convolution

Predict then Propagate: Graph Neural Networks meet Personalized PageRank

Learning Implicitly Recurrent CNNs Through Parameter Sharing