arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.13789cs.LOcs.AIcs.FLcs.GT
收藏

马尔可夫决策过程中概率目标与非概率目标的混合

Mixing Probabilistic and non-Probabilistic Objectives in Markov Decision Processes

University of Antwerp(安特卫普大学)

Raphaël Berthon, Shibashis Guha, Jean-François Raskin

摘要:

在本文中,我们考虑的算法,以确定存在的MDP的目标的布尔组合的战略。这些目标是Ω正则性质,需要强制执行,或者肯定,几乎肯定,存在,或非零概率。在这种情况下,相关的策略是随机无限记忆策略:无限记忆和随机化都可能需要最佳发挥。我们提供的算法来解决一般情况下的布尔组合,我们也调查相关的子情况。我们进一步报告这些问题的复杂性界限。

英文摘要:

In this paper, we consider algorithms to decide the existence of strategies in MDPs for Boolean combinations of objectives. These objectives are omega-regular properties that need to be enforced either surely, almost surely, existentially, or with non-zero probability. In this setting, relevant strategies are randomized infinite memory strategies: both infinite memory and randomization may be needed to play optimally. We provide algorithms to solve the general case of Boolean combinations and we also investigate relevant subcases. We further report on complexity bounds for these problems.

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。

↑