SOTAVerified

Dream to Control: Learning Behaviors by Latent Imagination

2019-12-03ICLR 2020Code Available1· sign in to hype

Danijar Hafner, Timothy Lillicrap, Jimmy Ba, Mohammad Norouzi

Code Available — Be the first to reproduce this paper.

Reproduce

Code

Abstract

Learned world models summarize an agent's experience to facilitate learning complex behaviors. While learning world models from high-dimensional sensory inputs is becoming feasible through deep learning, there are many potential ways for deriving behaviors from them. We present Dreamer, a reinforcement learning agent that solves long-horizon tasks from images purely by latent imagination. We efficiently learn behaviors by propagating analytic gradients of learned state values back through trajectories imagined in the compact state space of a learned world model. On 20 challenging visual control tasks, Dreamer exceeds existing approaches in data-efficiency, computation time, and final performance.

Tasks

Reproductions