On Context Distribution Shift in Task Representation Learning for Offline Meta RL

TL;DR

This paper addresses the challenge of context distribution shift in offline meta reinforcement learning by proposing a hard-sampling strategy to improve task representation robustness, leading to better performance on continuous control tasks.

Contribution

It introduces a novel hard-sampling-based training strategy for context encoders to mitigate distribution shift in offline meta RL.

Findings

Enhanced robustness of task representations with the proposed method.

Improved testing performance in continuous control tasks.

Better accumulated returns compared to baseline methods.

Abstract

Offline Meta Reinforcement Learning (OMRL) aims to learn transferable knowledge from offline datasets to enhance the learning process for new target tasks. Context-based Reinforcement Learning (RL) adopts a context encoder to expediently adapt the agent to new tasks by inferring the task representation, and then adjusting the policy based on this inferred representation. In this work, we focus on context-based OMRL, specifically on the challenge of learning task representation for OMRL. We conduct experiments that demonstrate that the context encoder trained on offline datasets might encounter distribution shift between the contexts used for training and testing. To overcome this problem, we present a hard-sampling-based strategy to train a robust task context encoder. Our experimental findings on diverse continuous control tasks reveal that utilizing our approach yields more robust task representations and better testing performance in terms of accumulated returns compared to baseline methods. Our code is available at https://github.com/ZJLAB-AMMI/HS-OMRL.