| Optimal Response Learning and Its Convergence in Multiagent Domains |
| |
| 作者姓名： | 张化祥 黄上腾 乐嘉锦 |
| |
| 作者单位： | [1]Department of Computer Science ＆ Engineering, Shanghai Jiaotong University, Shanghai 200030 [2]Department of Computer Science ＆ Engineering, Donghua University, Shanghai 200051 |
| |
| 摘 要： | In multiagent reinforcement learning, with different assumptions of the opponents＇ policies, an agent adopts quite different learning rules, and gets different learning performances. We prove that, in multiagent domains, convergence of the Q values is guaranteed only when an agent behaves optimally and its opponents＇ strategies satisfy certain conditions, and an agent can get best learning performances when it adopts the same learning algorithm as that of its opponents.
|
| 关 键 词： | Agent 马尔可夫决策过程 加强学习 学习规则 多主体域 |
| 收稿时间： | 2003-12-30 |
Optimal Response Learning and Its Convergence in Multiagent Domains |
| |
| ZHANG Hua-xiang,HUANG Shang-teng,LE Jia-jin.Optimal Response Learning and Its Convergence in Multiagent Domains[J].Journal of Donghua University,2005,22(3):116-119. |
| |
| Authors: | ZHANG Hua-xiang HUANG Shang-teng LE Jia-jin |
| |
| Abstract: | In multiagent reinforcement learning, with different assumptions of the opponents' policies, an agent adopts quite different learning rules, and gets different learning performances. We prove that, in multiagent domains, convergence of the Q values is guaranteed only when an agent behaves optimally and its opponents' strategies satisfy certain conditions, and an agent can get best learning performances when it adopts the same learning algorithm as that of its opponents. |
| |
| Keywords: | multiagent learning policy |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
