| Agent组织规则的再励学习 |
| |
| 引用本文： | 张伟,王一川,石纯一.Agent组织规则的再励学习[J].计算机研究与发展,2003,40(3):430-434. |
| |
| 作者姓名： | 张伟 王一川 石纯一 |
| |
| 作者单位： | 1. 烟台大学计算机科学与技术学院,烟台,264005;清华大学计算机科学与技术系,北京,100084
2. 清华大学计算机科学与技术系,北京,100084 |
| |
| 基金项目： | 国家自然科学基金 ( 69973 0 2 3 ,60 173 0 11)，国家“八六三”高技术研究发展计划基金 ( 2 0 0 1AA113 12 0 ) |
| |
| 摘 要： | Agent组织是一种灵活有效的多Agent系统求解方式。Agent组织规则在Agent组织的求解过程中起着重要作用，可以有效地减少冲突提高求解效率。给出了一种基于再励学习的Agent组织规则生成机制和相应的算法，通过实验表明了算法的有效性，改进了Zambonelli和Jennings等人关于Agent组织规则的工作。
|
| 关 键 词： | Agent Agent组织 Agent组织规则 再励学习 |
Agent Organization Rule Generation Based on Reinforcement Learning |
| |
| ZHANG Wei ,WANG Yi Chuan ,and SHI Chun Yi.Agent Organization Rule Generation Based on Reinforcement Learning[J].Journal of Computer Research and Development,2003,40(3):430-434. |
| |
| Authors: | ZHANG Wei WANG Yi Chuan and SHI Chun Yi |
| |
| Affiliation: | ZHANG Wei 1,2,WANG Yi Chuan 2,and SHI Chun Yi 2 1 |
| |
| Abstract: | Agent organization is a flexible and effective way of solving multi agent system problems Within an agent organization, the rule plays an important role for reducing conflicts and improving resolution efficiency In this paper, mechanisms and the associated algorithms of agent organization rule generation based on reinforcement learning are given The experiment shows that the algorithms are more effective than the normal method The agent organization rule generation method proposed improves the work conducted by Zambonelli and Jennings |
| |
| Keywords: | agent agent organization agent organization rule reinforcement learning |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
