Interactive Data Visualization in the browser, from Python

Grounded SCAN data set.

Code for "Recurrent Independent Mechanisms"

PyTorch implementations of Deep Reinforcement Learning algorithms (DQN, DDQN, A2C, VPG, TRPO, PPO, DDPG, TD3, SAC, ASAC, TAC, ATAC)

​TextWorld is a sandbox learning environment for the training and evaluation of reinforcement learning (RL) agents on text-based games.

Compositional generalization through meta sequence-to-sequence learning

PyTorch code to run synthetic experiments.

OpenDialKG: Explainable Conversational Reasoning with Attention-based Walks over Knowledge Graphs

Geometric Deep Learning Extension Library for PyTorch

Emergent communication in agents learning about object affordances

This repository provides code for reproducing experiments of the paper Talk The Walk: Navigating New York City Through Grounded Dialogue by Harm de Vries, Kurt Shuster, Dhruv Batra, Devi Parikh, Ja…

Word-order biases in deep-agent emergent communication

Measuring compositionality in representation learning

This repository provides text game simulators for research purposes.

On the pitfalls of measuring emergent communication

Coin collector game in Microsoft TextWorld, and a simple RL agent solving it.

Mind-aware Multi-agent Management Reinforcement Learning

Originally material for the AI4good summer school, now just a collection of stuff.

Code for "Learning Inductive Biases with Simple Neural Networks" (Feinman & Lake, 2018).

rllab is a framework for developing and evaluating reinforcement learning algorithms, fully compatible with OpenAI Gym.

Simple language-driven navigation tasks for studying compositional learning

Various tutorials given for welcoming new students at MILA.

Code for Emergent Translation in Multi-Agent Communication