Guiding Online Reinforcement Learning with Action-Free Offline Pretraining

TL;DR

This paper introduces AF-Guide, a novel method that leverages action-free offline datasets to enhance online reinforcement learning, improving sample efficiency and performance without requiring action logs during data collection.

Contribution

The paper proposes AF-Guide, combining an Action-Free Decision Transformer and Guided SAC to utilize action-free offline data for guiding online RL training, a novel approach in the field.

Findings

AF-Guide improves sample efficiency in online RL.

Using action-free offline data enhances performance.

The method is effective across different environments.

Abstract

Offline RL methods have been shown to reduce the need for environment interaction by training agents using offline collected episodes. However, these methods typically require action information to be logged during data collection, which can be difficult or even impossible in some practical cases. In this paper, we investigate the potential of using action-free offline datasets to improve online reinforcement learning, name this problem Reinforcement Learning with Action-Free Offline Pretraining (AFP-RL). We introduce Action-Free Guide (AF-Guide), a method that guides online training by extracting knowledge from action-free offline datasets. AF-Guide consists of an Action-Free Decision Transformer (AFDT) implementing a variant of Upside-Down Reinforcement Learning. It learns to plan the next states from the offline dataset, and a Guided Soft Actor-Critic (Guided SAC) that learns online with guidance from AFDT. Experimental results show that AF-Guide can improve sample efficiency and performance in online training thanks to the knowledge from the action-free offline dataset. Code is available at https://github.com/Vision-CAIR/AF-Guide.