Is Curiosity All You Need? On the Utility of Emergent Behaviours from Curious Exploration
This work addresses the challenge of leveraging curiosity-based learning for skill retention in reinforcement learning, offering an incremental improvement over existing methods.
The paper tackles the problem of emergent behaviors being lost during curiosity-driven exploration, proposing to retain these behaviors as reusable skills for related tasks. Experiments show that a simple policy snapshot method effectively reuses discovered behaviors for transfer tasks.
Curiosity-based reward schemes can present powerful exploration mechanisms which facilitate the discovery of solutions for complex, sparse or long-horizon tasks. However, as the agent learns to reach previously unexplored spaces and the objective adapts to reward new areas, many behaviours emerge only to disappear due to being overwritten by the constantly shifting objective. We argue that merely using curiosity for fast environment exploration or as a bonus reward for a specific task does not harness the full potential of this technique and misses useful skills. Instead, we propose to shift the focus towards retaining the behaviours which emerge during curiosity-based learning. We posit that these self-discovered behaviours serve as valuable skills in an agent's repertoire to solve related tasks. Our experiments demonstrate the continuous shift in behaviour throughout training and the benefits of a simple policy snapshot method to reuse discovered behaviour for transfer tasks.