arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2101.02230cs.LGcs.AI
收藏

学习动态感知的状态嵌入用于迁移学习

Learn Dynamic-Aware State Embedding for Transfer Learning

University College London(伦敦大学学院)

Kaige Yang

摘要:

迁移强化学习旨在通过利用从先前任务中获得的经验来提高解决看不见的新任务的样本效率。我们考虑的设置中,所有的任务(MDP)共享相同的环境动态的奖励功能。在这种情况下,MDP动态是一个很好的知识转移,这可以推断出均匀随机策略。然而,均匀随机策略产生的轨迹对策略改进没有帮助,严重影响了样本效率。相反,我们观察到,二进制MDP动态可以推断出任何政策,避免了需要统一的随机政策的轨迹。由于二进制MDP动态包含在所有任务上共享的状态结构,因此我们认为它适合传输。基于这一观察,我们引入了一种方法来推断二进制MDP动态在线,并在同一时间利用它来指导状态嵌入学习,然后转移到新的任务。我们将状态嵌入学习和策略学习分开。因此,学习状态嵌入是任务和策略不可知的,这使得它非常适合迁移学习。此外,为了便于探索的状态空间,我们提出了一种新的内在奖励的基础上推断的二进制MDP动态。我们的方法可以与无模型RL算法结合使用。我们在\algo{DQN}和\algo{A2C}的基础上给出了两个实例。大量实验结果表明,该方法在各种迁移学习任务中具有优势。

英文摘要:

Transfer reinforcement learning aims to improve the sample efficiency of solving unseen new tasks by leveraging experiences obtained from previous tasks. We consider the setting where all tasks (MDPs) share the same environment dynamic except reward function. In this setting, the MDP dynamic is a good knowledge to transfer, which can be inferred by uniformly random policy. However, trajectories generated by uniform random policy are not useful for policy improvement, which impairs the sample efficiency severely. Instead, we observe that the binary MDP dynamic can be inferred from trajectories of any policy which avoids the need of uniform random policy. As the binary MDP dynamic contains the state structure shared over all tasks we believe it is suitable to transfer. Built on this observation, we introduce a method to infer the binary MDP dynamic on-line and at the same time utilize it to guide state embedding learning, which is then transferred to new tasks. We keep state embedding learning and policy learning separately. As a result, the learned state embedding is task and policy agnostic which makes it ideal for transfer learning. In addition, to facilitate the exploration over the state space, we propose a novel intrinsic reward based on the inferred binary MDP dynamic. Our method can be used out-of-box in combination with model-free RL algorithms. We show two instances on the basis of \algo{DQN} and \algo{A2C}. Empirical results of intensive experiments show the advantage of our proposed method in various transfer learning tasks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑