arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.09273cs.LG
收藏

基于最大似然模型转移的强化学习

Reinforcement Learning in the Wild with Maximum Likelihood-based Model Transfer

Zenseact;Chalmers University of Technology(查尔姆斯理工大学);Scool;CRIStAL(里尔信号与自动化信息研究中心);CNRS(法国国家科学研究中心);University of Oslo(奥斯陆大学);University of Neuchatel(纳沙泰尔大学)

Hannes Eriksson, Debabrota Basu, Tommy Tram, Mina Alibeigi, Christos Dimitrakakis

摘要:

在本文中,我们研究的问题,转移现有的马尔可夫决策过程(MDP)模型,学习和规划有效地在一个未知的,但类似的MDP。我们称之为\textit{Model Transfer Reinforcement Learning (MTRL)}问题。首先,我们制定MTRL离散MDP和线性二次调节器(LQR)连续状态行动。然后,我们提出了一个通用的两阶段算法,MLEMTRL,以解决离散和连续设置的MTRL问题。在第一阶段中,MLEMTRL使用基于\textit{constrained Maximum Likelihood Estimation (MLE)}的方法来使用一组已知的MDP模型来估计目标MDP模型。在第二阶段,使用估计的目标MDP模型,MLEMTRL部署一个基于模型的规划算法适合于MDP类。从理论上讲,我们证明了最坏情况下的遗憾界限MLEMTRL在可实现和不可实现的设置。我们的经验表明,MLEMTRL允许更快地学习新的MDP比从头开始学习,并达到接近最佳的性能取决于可用的MDP和目标MDP的相似性。

英文摘要:

In this paper, we study the problem of transferring the available Markov Decision Process (MDP) models to learn and plan efficiently in an unknown but similar MDP. We refer to it as \textit{Model Transfer Reinforcement Learning (MTRL)} problem. First, we formulate MTRL for discrete MDPs and Linear Quadratic Regulators (LQRs) with continuous state actions. Then, we propose a generic two-stage algorithm, MLEMTRL, to address the MTRL problem in discrete and continuous settings. In the first stage, MLEMTRL uses a \textit{constrained Maximum Likelihood Estimation (MLE)}-based approach to estimate the target MDP model using a set of known MDP models. In the second stage, using the estimated target MDP model, MLEMTRL deploys a model-based planning algorithm appropriate for the MDP class. Theoretically, we prove worst-case regret bounds for MLEMTRL both in realisable and non-realisable settings. We empirically demonstrate that MLEMTRL allows faster learning in new MDPs than learning from scratch and achieves near-optimal performance depending on the similarity of the available MDPs and the target MDP.

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。

↑