| 预序马尔柯夫决策规划 |
| |
| 引用本文： | 吴吉山.预序马尔柯夫决策规划[J].华中科技大学学报(自然科学版),1987(Z3). |
| |
| 作者姓名： | 吴吉山 |
| |
| 作者单位： | 江西冶金学院 |
| |
| 摘 要： | 本文在文献1]～3]的基础上,建立了一般意义下的预序模型,并研究了该模型最优策略的结构。文中彻底放弃了状态转移是确定性的假设,将策略从确定性策略类Π~d放宽到一般的随机策略类Π上进行讨论,从而大大地推广了文献4]的结果。
|
| 关 键 词： | 马尔柯夫决策 预序模型 最优策略 |
Ordinal Markov Decision Programming |
| |
| Wu Jishan.Ordinal Markov Decision Programming[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,1987(Z3). |
| |
| Authors: | Wu Jishan |
| |
| Abstract: | Based on 1]-3], this paper presents the ordinal MDp model in a broad sense. The structure of optimal policy for this model is discussed. The assumption that the state transition is deterministic is thoroughly given up and the policy is extended from a deterministic one (d) to an ordinary stochastic one () , leading to an extension of the results given in 4]. |
| |
| Keywords: | Markovian decision MDP model Optimal policy |
| 本文献已被 CNKI 等数据库收录！ |
