Bisimulation Makes Analogies in Goal-Conditioned Reinforcement Learning

June 09, 2022


Building generalizable goal-conditioned agents from rich observations is a key to reinforcement learning (RL) solving real world problems. Traditionally in goal-conditioned RL, an agent is provided with the exact goal they intend to reach. However, it is often not realistic to know the configuration of the goal before performing a task. A more scalable framework would allow us to provide the agent with an example of an analogous task, and have the agent then infer what the goal should be for its current state. We propose a new form of state abstraction called goal-conditioned bisimulation that captures functional equivariance, allowing for the reuse of skills to achieve new goals. We learn this representation using a metric form of this abstraction, and show its ability to generalize to new goals in simulation manipulation tasks. Further, we prove that this learned representation is sufficient not only for goal conditioned tasks, but is amenable to any downstream task described by a state-only reward function.

Download the Paper


Written by

Amy Zhang

Ashvin Nair

Patrick Yin

Philippe Hansen-Estruch

Sergey Levine



Research Topics

Reinforcement Learning

Help Us Pioneer The Future of AI

We share our open source frameworks, tools, libraries, and models for everything from research exploration to large-scale production deployment.