The LP Approach In Average Reward MDPs With Multiple Cost Constraints : The Unbounded Case

スポンサーリンク

概要

• 論文の詳細を見る

• In this paper we study constrained Markov decision processes with countable states and unbounded reward. A corresponding linear programming and its dual are formulated. Under some reasonable conditions we prove thier solvability and absence of a duality gap. By the convexity assumption we derive the existence of a constrained optimal stationary policy.

• 千葉大学の論文
• 1997-02-28

著者

• Kurano Masami

  千葉大学教育学部

• Huang Youqiang

  千葉大学教育学部

関連論文

• The LP Approach In Average Reward MDPs With Multiple Cost Constraints : The Unbounded Case

スポンサーリンク