Transfer Reinforcement Learning for Differing Action Spaces via Q-Network Representations
This addresses a niche problem in reinforcement learning for researchers, focusing on an incremental advancement in transfer learning for action space differences.
The paper tackles transfer reinforcement learning between tasks with different action spaces by proposing a reward shaping method based on source embedding similarity, tested on Acrobot-v1 and Pendulum-v0 domains; results show improvement in discrete action spaces but not in continuous ones compared to baselines.
Transfer learning approaches in reinforcement learning aim to assist agents in learning their target domains by leveraging the knowledge learned from other agents that have been trained on similar source domains. For example, recent research focus within this space has been placed on knowledge transfer between tasks that have different transition dynamics and reward functions; however, little focus has been placed on knowledge transfer between tasks that have different action spaces. In this paper, we approach the task of transfer learning between domains that differ in action spaces. We present a reward shaping method based on source embedding similarity that is applicable to domains with both discrete and continuous action spaces. The efficacy of our approach is evaluated on transfer to restricted action spaces in the Acrobot-v1 and Pendulum-v0 domains. A comparison with two baselines shows that our method does not outperform these baselines in these continuous action spaces but does show an improvement in these discrete action spaces. We conclude our analysis with future directions for this work.