计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (30): 24-25.DOI: 10.3778/j.issn.1002-8331.2008.30.007

• 博士论坛 • 上一篇    下一篇

一种优化的基于增强学习协商策略

孙天昊,朱庆生,李双庆,周明强   

  1. 重庆大学 计算机学院,重庆 400030
  • 收稿日期:2008-07-01 修回日期:2008-07-28 出版日期:2008-10-21 发布日期:2008-10-21
  • 通讯作者: 孙天昊

Optimized negotiation strategy based on reinforcement learning

SUN Tian-hao,ZHU Qing-sheng,LI Shuang-qing,ZHOU Ming-qiang   

  1. College of Computer,Chongqing University,Chongqing 400030,China
  • Received:2008-07-01 Revised:2008-07-28 Online:2008-10-21 Published:2008-10-21
  • Contact: SUN Tian-hao

摘要: 增强学习可以帮助协商Agent选择最优行动实现其最终目标。对基于增强学习的协商策略进行优化,在协商过程中充分利用对手的历史信息,加快协商解的收敛和提高协商解的质量。最后通过实验验证了算法的有效性和可用性。

关键词: 增强学习, 协商策略, 协商历史

Abstract: Negotiation agent can use reinforcement learning to select its best actions and reach its final goal.This paper proposes an optimized negotiation strategy based on reinforcement learning.In the middle of negotiation process,it makes the best use of the opponent’s negotiation history,in order to quicken the negotiation result convergence and enhance the negotiation result quality.Finally,the algorithm is proved to be effective and practical by experiment.

Key words: reinforcement learning, negotiation strategy, negotiation history