arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1301.0601cs.LGstat.ML
收藏

部分已知世界动态的强化学习

Reinforcement Learning with Partially Known World Dynamics

Stanford University(斯坦福大学)

Christian R. Shelton

摘要:

如果建模者能够将领域知识传授给算法,强化学习将在现实世界的问题上取得更好的成功。大多数问题都有隐藏的状态和未知的动态。部分可观测马尔可夫决策过程(POMDPs)允许两者的建模。不幸的是,他们没有提供一个自然的框架,在其中指定的域动态知识。设计者必须要么承认对动态一无所知,要么完全指定动态(从而将其转化为规划问题)。我们提出了一个新的框架称为部分已知马尔可夫决策过程(PKMDP),它允许设计者指定已知的动态,而仍然留下部分的环境的动态未知。该模型不仅表示环境的动态,但也代理的动态知识。我们针对该模型提出了一种基于重要性采样的强化学习算法。该算法结合规划的基础上已知的动态和学习的未知动态。我们的研究结果清楚地表明了添加领域知识的能力以及由此产生的学习益处。

英文摘要:

Reinforcement learning would enjoy better success on real-world problems if domain knowledge could be imparted to the algorithm by the modelers. Most problems have both hidden state and unknown dynamics. Partially observable Markov decision processes (POMDPs) allow for the modeling of both. Unfortunately, they do not provide a natural framework in which to specify knowledge about the domain dynamics. The designer must either admit to knowing nothing about the dynamics or completely specify the dynamics (thereby turning it into a planning problem). We propose a new framework called a partially known Markov decision process (PKMDP) which allows the designer to specify known dynamics while still leaving portions of the environment s dynamics unknown.The model represents NOT ONLY the environment dynamics but also the agents knowledge of the dynamics. We present a reinforcement learning algorithm for this model based on importance sampling. The algorithm incorporates planning based on the known dynamics and learning about the unknown dynamics. Our results clearly demonstrate the ability to add domain knowledge and the resulting benefits for learning.

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。

↑