SkillS: Adaptive Skill Sequencing for Efficient Temporally-Extended Exploration

TL;DR

SkillS introduces an adaptive skill sequencing method that enhances exploration efficiency in reinforcement learning by learning to sequence existing skills and directly deriving the final policy from raw experience, outperforming classical methods.

Contribution

The paper proposes a novel approach that sequences existing skills for exploration and learns the final policy directly, addressing limitations of prior skill transfer methods.

Findings

Outperforms classical skill transfer methods across multiple tasks

Enables rapid adaptation and efficient data collection

Highlights importance of components through ablation studies

Abstract

The ability to effectively reuse prior knowledge is a key requirement when building general and flexible Reinforcement Learning (RL) agents. Skill reuse is one of the most common approaches, but current methods have considerable limitations.For example, fine-tuning an existing policy frequently fails, as the policy can degrade rapidly early in training. In a similar vein, distillation of expert behavior can lead to poor results when given sub-optimal experts. We compare several common approaches for skill transfer on multiple domains including changes in task and system dynamics. We identify how existing methods can fail and introduce an alternative approach to mitigate these problems. Our approach learns to sequence existing temporally-extended skills for exploration but learns the final policy directly from the raw experience. This conceptual split enables rapid adaptation and thus efficient data collection but without constraining the final solution.It significantly outperforms many classical methods across a suite of evaluation tasks and we use a broad set of ablations to highlight the importance of differentc omponents of our method.