Ensemble Successor Representations for Task Generalization in Offline-to-Online Reinforcement Learning

TL;DR

This paper proposes an ensemble successor representation approach for efficient task generalization in offline-to-online reinforcement learning, enabling rapid adaptation to new tasks using offline data.

Contribution

It introduces a novel ensemble successor representation method that improves task generalization and online adaptation in offline-to-online RL settings.

Findings

Outperforms existing methods in generalizing to unseen tasks

Enables faster online adaptation with offline data

Robust to datasets with different coverage

Abstract

In Reinforcement Learning (RL), training a policy from scratch with online experiences can be inefficient because of the difficulties in exploration. Recently, offline RL provides a promising solution by giving an initialized offline policy, which can be refined through online interactions. However, existing approaches primarily perform offline and online learning in the same task, without considering the task generalization problem in offline-to-online adaptation. In real-world applications, it is common that we only have an offline dataset from a specific task while aiming for fast online-adaptation for several tasks. To address this problem, our work builds upon the investigation of successor representations for task generalization in online RL and extends the framework to incorporate offline-to-online learning. We demonstrate that the conventional paradigm using successor features cannot effectively utilize offline data and improve the performance for the new task by online fine-tuning. To mitigate this, we introduce a novel methodology that leverages offline data to acquire an ensemble of successor representations and subsequently constructs ensemble Q functions. This approach enables robust representation learning from datasets with different coverage and facilitates fast adaption of Q functions towards new tasks during the online fine-tuning phase. Extensive empirical evaluations provide compelling evidence showcasing the superior performance of our method in generalizing to diverse or even unseen tasks.