Task-agnostic Lifelong Robot Learning with Retrieval-based Weighted Local Adaptation

TL;DR

This paper introduces a retrieval-based local adaptation method for lifelong robot learning that effectively mitigates catastrophic forgetting in task-free scenarios by selectively restoring relevant knowledge without relying on task IDs.

Contribution

It proposes a novel retrieval-based local adaptation technique combined with experience replay and a selective weighting mechanism for task-free lifelong robot learning.

Findings

Improved performance in diverse manipulation tasks

Effective knowledge restoration without task boundaries

Scalable approach for open-ended lifelong learning

Abstract

A fundamental objective in intelligent robotics is to move towards lifelong learning robot that can learn and adapt to unseen scenarios over time. However, continually learning new tasks would introduce catastrophic forgetting problems due to data distribution shifts. To mitigate this, we store a subset of data from previous tasks and utilize it in two manners: leveraging experience replay to retain learned skills and applying a novel Retrieval-based Local Adaptation technique to restore relevant knowledge. Since a lifelong learning robot must operate in task-free scenarios, where task IDs and even boundaries are not available, our method performs effectively without relying on such information. We also incorporate a selective weighting mechanism to focus on the most "forgotten" skill segment, ensuring effective knowledge restoration. Experimental results across diverse manipulation tasks demonstrate that our framework provides a scalable paradigm for lifelong learning, enhancing robot performance in open-ended, task-free scenarios.