| 报酬无界的平均准则马氏决策过程 |
| |
| 引用本文： | 胡奇英.报酬无界的平均准则马氏决策过程[J].运筹学学报,2002,6(1):1-8. |
| |
| 作者姓名： | 胡奇英 |
| |
| 作者单位： | 西安电子科技大学经济管理学院,西安,710071 |
| |
| 基金项目： | The project was supported by the National Natural Science Foundation of China. |
| |
| 摘 要： | 本文对可数状态集、非空决策集、报酬无界的平均准则马氏决策过程，提出了一组新的条件，在此条件下存在（ε）最优平稳策略，且当最优不等式中的和有定义时最优不等式也成立。
|
| 关 键 词： | 马氏决策过程 平均准则最优不等式 无界报酬 非空决策集 |
Average Optimality in Markov Decision Processes with Unbounded Rewards |
| |
| QIYING HU.Average Optimality in Markov Decision Processes with Unbounded Rewards[J].OR Transactions,2002,6(1):1-8. |
| |
| Authors: | QIYING HU |
| |
| Abstract: | This paper studies average optimality in Markov decision processes with countablestate space, nonempty action sets and unbounded reward function. New conditions arediscussed under which there exists an (ε) optimal stationary policy, and that the averagecriterion optimality inequality holds when the summation in it is well defined. |
| |
| Keywords: | Markov decision process average criterion optimality inequality un-bounded rewards nonempty action sets |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
