arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1402.6763math.OCcs.AIcs.NAmath.NA
收藏

大规模马尔可夫决策问题的线性规划

Linear Programming for Large-Scale Markov Decision Problems

Queensland University of Technology(昆士兰科技大学);UC Berkeley(加州大学伯克利分校);QUT(昆士兰科技大学)

Yasin Abbasi-Yadkori, Peter L. Bartlett, Alan Malek

AI总结:

本文提出通过低维策略集与线性规划对大规模马尔可夫决策过程进行优化,通过随机凸优化和约束采样技术实现性能逼近,不依赖状态空间大小。

摘要:

我们考虑如何控制具有大状态空间的马尔可夫决策过程(MDP)以最小化平均成本。由于对于大规模问题无法与最优策略竞争,我们追求更现实的目标,即与低维策略集竞争。我们使用MDP平均成本问题的对偶线性规划形式,其中变量是状态-动作对的平稳分布,并考虑一个低维子集的平稳分布邻域(以状态-动作特征定义)作为比较类。我们提出两种技术,一种基于随机凸优化,另一种基于约束采样。在两种情况下,我们给出的界限表明,我们的算法性能可以逼近比较类中任何策略的最佳性能。最重要的是,这些结果依赖于比较类的大小,而不是状态空间的大小。初步实验显示,所提出算法在排队应用中有效。

英文摘要:

We consider the problem of controlling a Markov decision process (MDP) with a large state space, so as to minimize average cost. Since it is intractable to compete with the optimal policy for large scale problems, we pursue the more modest goal of competing with a low-dimensional family of policies. We use the dual linear programming formulation of the MDP average cost problem, in which the variable is a stationary distribution over state-action pairs, and we consider a neighborhood of a low-dimensional subset of the set of stationary distributions (defined in terms of state-action features) as the comparison class. We propose two techniques, one based on stochastic convex optimization, and one based on constraint sampling. In both cases, we give bounds that show that the performance of our algorithms approaches the best achievable by any policy in the comparison class. Most importantly, these results depend on the size of the comparison class, but not on the size of the state space. Preliminary experiments show the effectiveness of the proposed algorithms in a queuing application.

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。

↑