arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1302.2552cs.LG
收藏

强化学习中状态表示的选择

Selecting the State-Representation in Reinforcement Learning

INRIA Lille(法国国家信息与自动化研究所里尔中心)

Odalric-Ambrym Maillard, Rémi Munos, Daniil Ryabko

摘要:

研究了强化学习中状态表示的选择问题。给出了观测的几种模型(将过去的观测映射到有限集的函数),并且已知对于这些模型中的至少一个,所得到的状态动态确实是马尔可夫的。在既不知道哪一个模型是正确的模型,也不知道所得到的MDP的概率特性是什么的情况下,需要获得与正确模型的最优策略一样多的奖励(或者如果有几个正确模型,则为最好的模型)。我们提出了一个算法,实现了这一点,遗憾的顺序T^{2/3},其中T是地平线时间。

英文摘要:

The problem of selecting the right state-representation in a reinforcement learning problem is considered. Several models (functions mapping past observations to a finite set) of the observations are given, and it is known that for at least one of these models the resulting state dynamics are indeed Markovian. Without knowing neither which of the models is the correct one, nor what are the probabilistic characteristics of the resulting MDP, it is required to obtain as much reward as the optimal policy for the correct model (or for the best of the correct models, if there are several). We propose an algorithm that achieves that, with a regret of order T^{2/3} where T is the horizon time.

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。

↑