| 基于状态软集结的相对值迭代算法 |
| |
| 引用本文： | 胡光华,吴沧浦,乔治·瑟彬珂.基于状态软集结的相对值迭代算法[J].控制理论与应用,2000,17(3):415-418. |
| |
| 作者姓名： | 胡光华 吴沧浦 乔治·瑟彬珂 |
| |
| 作者单位： | 1. 北京理工大学自动控制系北京,100081
2. 达特茅斯学院工学院,新罕尔布什州汉诺威,NH03755,美国 |
| |
| 基金项目： | Foundationitem :supportedbytheNationalNaturalScienceFoundationofChina (696740 0 5) . |
| |
| 摘 要： | 在大规模随机控制问题中，值函数逼近是一种克服维数灾的方法，考虑平均模型马氏决策规划（ＭＤＰ）的状态软集结相对值迭代算法，在Ｓｐａｎ压缩的条件下，证明了该算法的收敛性，同时还给出了其误差估计。
|
| 关 键 词： | 随机控制 状态软集结 相对值 迭代算法 |
Relative Value Iteration Algorithm with Soft State Aggregation |
| |
| HU Guanghua,WU Cangpu,George Cybenko.Relative Value Iteration Algorithm with Soft State Aggregation[J].Control Theory & Applications,2000,17(3):415-418. |
| |
| Authors: | HU Guanghua WU Cangpu George Cybenko |
| |
| Abstract: | A straightforward way to dispel the curse of dimensionality in large stochastic control problems is to replace the lookup table with a generalized function approximator such as state aggregation. The relative value iteration algorithm for average reward Markov decision processes (MDP) with soft state aggregation is investigated. Under a condition of the contraction with span semi norm, the convergence of the proposed algorithm is proved and an error bound of the approximation is also given. |
| |
| Keywords: | dynamic programming Markov decision processes compact representation state aggregation average reward |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《控制理论与应用》浏览原始摘要信息 |
