May 31, 2017

Modout: Learning Multi-modal Architectures by Stochastic Regularization

IEEE Conference on Automatic Face and Gesture Recognition (FG 2017)

This paper describes Modout, a model selection method based on stochastic regularization, which is particularly useful in the multi-modal setting.

Fan Li, Natalia Neverova, Christian Wolf, Graham Taylor
April 26, 2017

A New AI Evaluation Cosmos: Ready to Play the Game?

AI Magazine

We report on a series of new platforms and events dealing with AI evaluation that may change the way in which AI systems are compared and their progress is measured.

José Hernández-Orallo, Marco Baroni, Jordi Bieger, Nader Chmait, David L. Dowe, Katja Hofmann, Fernando Martínez-Plumed, Claes Strannegård, Kristinn R. Thórisson
April 24, 2017

Towards Principled Methods for Training Generative Adversarial Networks

International Conference on Learning Representations (ICLR) 2017

The goal of this paper is not to introduce a single algorithm or method, but to make theoretical steps towards fully understanding the training dynamics of generative adversarial networks. I

Martin Arjovsky, Leon Bottou
April 24, 2017

Revisiting Classifier Two-Sample Tests for GAN Evaluation and Causal Discovery

International Conference on Learning Representations (ICLR) 2017

In this paper, we aim to revive interest in the use of binary classifiers for two-sample testing. To this end, we review their fundamentals, previous literature on their use, compare their performance against alternative state-of-the-art two-sample tests, and propose them to evaluate generative adversarial network models applied to image synthesis.

David Lopez-Paz, Maxime Oquab
April 24, 2017

Automatic Rule Extraction from Long Short Term Memory Networks

International Conference on Learning Representations (ICLR) 2017

In this paper we consider Long Short Term Memory networks (LSTMs) and demonstrate a new approach for tracking the importance of a given input to the LSTM for a given output.

W. James Murdoch, Arthur Szlam
April 24, 2017

Variable Computation in Recurrent Neural Networks

International Conference on Learning Representations (ICLR) 2017

In this paper, we explore a modification to existing recurrent units which allows them to learn to vary the amount of computation they perform at each step, without prior knowledge of the sequence’s time structure.

Yacine Jernite, Edouard Grave, Armand Joulin, Tomas Mikolov
April 24, 2017

Episodic Exploration for Deep Deterministic Policies for StarCraft Micro-Management

International Conference on Learning Representations (ICLR) 2017

We consider scenarios from the real-time strategy game StarCraft as benchmarks for reinforcement learning algorithms.

Nicolas Usunier∗, Gabriel Synnaeve∗, Zeming Lin, Soumith Chintala
April 24, 2017

Learning through Dialogue Interactions by Asking Questions

International Conference on Learning Representations (ICLR) 2017

In this work, we explore a dialogue agents ability to interact with users by both responding to questions and by asking questions, and importantly to learn from both types of interaction, by designing a simulator and a set of synthetic tasks in the movie domain that allow such interactions between a learner and a teacher.

Jiwei Li, Alexander H. Miller, Sumit Chopra, Marc’Aurelio Ranzato, Jason Weston
April 24, 2017

Dialogue Learning with Human-in-the-Loop

International Conference on Learning Representations (ICLR) 2017

n this paper we explore interacting with a dialogue partner in a reinforcement learning setting where the bot improves its question-answering ability from feedback a teacher gives following its generated responses.

Jiwei Li, Alexander H. Miller, Sumit Chopra, Marc’Aurelio Ranzato, Jason Weston
April 24, 2017

Training Agent for First-Person Shooter Game With Actor-Critic Curriculum Learning

International Conference on Learning Representations (ICLR) 2017

In this paper, we propose a new framework for training vision-based agent for First-Person Shooter (FPS) Game, in particular Doom.

Yuxin Wu, Yuandong Tian
April 24, 2017

Unsupervised Cross-Domain Image Generation

International Conference on Learning Representations (ICLR) 2017

We study the problem of transferring a sample in one domain to an analog sample in another domain. Given two related domains, S and T, we would like to learn a generative function G that maps an input sample from S to the domain T, such that the output of a given representation function f, which accepts inputs in either domains, would remain unchanged.

Yaniv Taigman, Adam Polyak, Lior Wolf
April 24, 2017

An Analytical Formula of Population Gradient for Two-Layered ReLU network and its Applications in Convergence and Critical Point Analysis

International Conference on Learning Representations (ICLR) 2017

In this paper, we explore theoretical properties of training a two-layered ReLU network g(x; w) = PK j=1 σ(w | j x) with centered d-dimensional spherical Gaussian input x (σ=ReLU). We train our network with gradient descent on w to mimic the output of a teacher network with the same architecture and fixed parameters w∗.

Yuandong Tian
April 24, 2017

Improving Neural Language Models with a Continuous Cache

International Conference on Learning Representations (ICLR) 2017

We propose an extension to neural network language models to adapt their prediction to the recent history. Our model is […]

Edouard Grave, Armand Joulin, Nicolas Usunier
April 24, 2017

CommAI: Evaluating the First Steps Towards a Useful General AI

ICLR 2017 Workshop

We propose a set of concrete desiderata for general AI, together with a platform to test machines on how well they satisfy such desiderata, while keeping all further complexities to a minimum.

Marco Baroni, Armand Joulin, Allan Jabri, German Kruszewski, Angeliki Lazaridou, Klemen Simonic, Tomas Mikolov
April 24, 2017

Learning End-to-End Goal-Oriented Dialog

International Conference on Learning Representations (ICLR) 2017

This paper proposes a testbed to break down the strengths and shortcomings of end-to-end dialog systems in goal-oriented applications.

Antoine Bordes, Y-Lan Boureau, Jason Weston
April 24, 2017

Multi-Agent Cooperation and the Emergence of (Natural) Language

International Conference on Learning Representations (ICLR) 2017

This paper proposes a framework for language learning that relies on multi-agent communication.

Angeliki Lazaridou, Alexander Peysakhovich, Marco Baroni
February 7, 2017

Exploring Normalization in Deep Residual Networks with Concatenated Rectified Linear Units

AAAI-17

This paper analyzes the role of Batch Normalization (BatchNorm) layers on ResNets in the hope of improving the current architecture and better incorporating other normalization techniques, such as Normalization Propagation (NormProp), into ResNets.

Wenling Shang, Justin Chiu, Kihyuk Sohn
December 6, 2016

Feedback Neural Network for Weakly Supervised Geo-Semantic Segmentation

Arxiv

We propose a novel neural network architecture to perform weakly-supervised learning by suppressing irrelevant neuron activations. When applied to a practical challenge of transforming satellite images into a map of settlements and individual buildings it delivers results that show superior performance and efficiency.

Xianming Liu, Amy Zhang, Tobias Tiecke, Andreas Gros, Thomas S. Huang
December 6, 2016

Population Density Estimation with Deconvolutional Neural Networks

Workshop on Large Scale Computer Vision at NIPS 2016

This work is part of the Internet.org initiative to provide connectivity all over the world. Population density data is helpful in driving a variety of technology decisions, but currently, a microscopic dataset of population doesn’t exist. Current state of the art population density datasets are at ~1000km2 resolution. To create a better dataset, we have obtained 1PB of satellite imagery at 50cm/pixel resolution to feed through our building classification pipeline.

Amy Zhang, Andreas Gros, Tobias Tiecke, Xianming Liu
November 30, 2016

Semantic Segmentation using Adversarial Networks

Workshop on Adversarial Training at NIPS 2016

Adversarial training has been shown to produce state of the art results for generative image modeling. In this paper we propose an adversarial training approach to train semantic segmentation models.

Pauline Luc, Camille Couprie, Soumith Chintala, Jakob Verbeek