Reinforcement Learning from Passive Data via Latent Intentions

TL;DR

This paper introduces a method for learning useful representations for reinforcement learning from passive observational data by modeling intentions, enabling RL agents to learn from unlabeled videos without explicit rewards or actions.

Contribution

The paper presents a novel approach that learns intentions from passive data using a temporal difference objective, allowing the extraction of features for downstream RL tasks without active interaction.

Findings

Learns representations that improve value prediction in downstream tasks.

Effective with diverse passive data, including videos from different embodiments.

Theoretically and empirically demonstrates the utility of intention modeling from passive data.

Abstract

Passive observational data, such as human videos, is abundant and rich in information, yet remains largely untapped by current RL methods. Perhaps surprisingly, we show that passive data, despite not having reward or action labels, can still be used to learn features that accelerate downstream RL. Our approach learns from passive data by modeling intentions: measuring how the likelihood of future outcomes change when the agent acts to achieve a particular task. We propose a temporal difference learning objective to learn about intentions, resulting in an algorithm similar to conventional RL, but which learns entirely from passive data. When optimizing this objective, our agent simultaneously learns representations of states, of policies, and of possible outcomes in an environment, all from raw observational data. Both theoretically and empirically, this scheme learns features amenable for value prediction for downstream tasks, and our experiments demonstrate the ability to learn from many forms of passive data, including cross-embodiment video data and YouTube videos.