Paper
Learning and Querying Fast Generative Models for Reinforcement Learning
A key challenge in model-based reinforcement learning (RL) is to synthesize computationally efficient and accurate environment models. We show that carefully designed generative models that learn and operate on compact state representations, so-called state-space models, substantially reduce the computational costs for predicting outcomes of sequences of actions. Extensive experiments establish that state-space models accurately capture the dynamics of Atari games from the Arcade Learning Environment from raw pixels. The computational speed-up of state-space models while maintaining high accuracy makes their application in RL feasible: We demonstrate that agents which query these models for decision making outperform strong model-free baselines on the game MSPACMAN, demonstrating the potential of using learned environment models for planning.
Authors: Buesing, Lars · Weber, Theophane · Racaniere, Sebastien · Eslami, S. M. Ali · Rezende, Danilo · Reichert, David P. · Viola, Fabio · Besse, Frederic · Gregor, Karol · Hassabis, Demis · Wierstra, Daan