基于Q-强化学习和Adaboost算法的自适应谈判方法  

Self-adaptive negotiation method based on Q-reinforcement learning and Adaboost algorithm

作　　者：庞婷 郭绍永 何喜军[2] 蒋国瑞[2] PANG Ting;GUO Shaoyong;HE Xijun;JIANG Guorui(Modern Education Technology Center,Xinxiang Medical University,Xinxiang 453003,China;Department of Economics and Management,Beijing University of Technology,Beijing 100124,China)

机构地区：[1]新乡医学院现代教育技术中心,新乡453003 [2]北京工业大学经济管理学院,北京100124

出　　处：《江苏科技大学学报（自然科学版）》2018年第4期564-568,共5页Journal of Jiangsu University of Science and Technology:Natural Science Edition

基　　金：国家自然科学基金资助项目(71371018);国家社会科学青年基金资助项目(13CGL002)

摘　　要：为有效提高谈判效率,增强谈判主体的自学习能力,文中提出一种自适应谈判方法.该方法设定一种让步谈判策略,采用Q-强化学习算法计算谈判主体的让步幅度,然后考虑对手行为,使用Adaboost算法预测对手提议而调整让步幅度.算例仿真结果表明,使用该方法减少了谈判次数,缩短了谈判时间,不易陷入局部最优,增强了自学习能力,提升了主体满意度,优化了谈判效果.To effectively improve the negotiation efficiency and enhance the learning ability of negotiation agent,a self-adaptive negotiation method is proposed.A concession negotiation strategy is set,and the Q-reinforcement learning algorithm is used to calculate the concession extent of negotiation agent.Then we consider the opponent′s behavior,uses the Adaboost algorithm to predict opponent′s proposal and adjust the concession extent.The simulation result of the example shows that the method can reduce the negotiation times and shorten the negotiation time,not easily fall into the local optimum and strengthen the self-learning ability,improve the agent satisfaction and optimize the negotiation effect.

关 键 词：让步谈判 Q-强化学习 ADABOOST算法 自适应方法 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]