| 基于增强学习的多agent自动协商研究 |
| |
| 引用本文： | 杨明,嘉莉,邱玉辉.基于增强学习的多agent自动协商研究[J].计算机工程与应用,2004,40(33):98-100,117. |
| |
| 作者姓名： | 杨明 嘉莉 邱玉辉 |
| |
| 作者单位： | 西南师范大学计算机与信息科学学院,重庆,400715 |
| |
| 摘 要： | 该文通过对协商协议的引入,对提议形式、协商流程的分析,结合多属性效用理论和连续决策过程,提出了一个开放的、动态的、支持学习机制的形式化多问题自动协商模型。并在模型的基础上分别对评估提议、更新信念、生成提议等协商过程作了详细描述;对传统Q学习进行了扩充,设计了基于agent的当前信念和最近探索盈余的动态Q学习算法。
|
| 关 键 词： | 增强学习 自动协商 Q学习 评估提议 |
| 文章编号： | 1002-8331-(2004)33-0098-03 |
Research on Automated Negotiation in Multi-agent System Based on Reinforcement Learning |
| |
| Yang Ming Jia Li,Qiu Yuhui.Research on Automated Negotiation in Multi-agent System Based on Reinforcement Learning[J].Computer Engineering and Applications,2004,40(33):98-100,117. |
| |
| Authors: | Yang Ming Jia Li Qiu Yuhui |
| |
| Abstract: | By presenting negotiation protocol and analyzing negotiation flow,based on multi-attribute utility theory and sequential decision-making process,this paper establishes an open and dynamic formalized negotiation model,which is embedded learning mechanism.This paper also describes in detail negotiation process based on the model,such as eval-uating offers,belief revision and proposing counteroffers.In addition,this paper restructures the traditional Q-learning into a dynamic Q-learning algorithm by introducing current beliefs and recent exploration bonus. |
| |
| Keywords: | reinforcement learning automated negotiation Q-learning evaluating offers |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
