基于遗忘理论和效用的多Agent多问题协商  

Multi-issue Negotiation Based on Forgetting Theory and Utility

作　　者：吴海燕[1] 胡山立[2] 

机构地区：[1]福建师范大学数学与计算机科学学院,福建福州350007 [2]福州大学数学与计算机科学学院,福建福州350002

出　　处：《广西师范大学学报（自然科学版）》2008年第1期162-165,共4页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家自然科学基金资助项目(60373079)

摘　　要：许多基于对策论的协商方法,倾向于假设Agent的效用固定不变,默认多问题之间是完全可以互相补偿的,但实际情况往往不是这样。在此提出基于遗忘理论的多问题协商,结合逻辑学和对策论的方法,达到了Agent协商逻辑理性和效用理性的平衡。在协商中,动态调整问题的取值,采用逻辑与来处理多Agent之间不可补偿的多问题协商,为多Agent的多问题协商提供新方法。Abstract .Many negotiation methods ,which are based on the game theory ,trend to suppose that the decision utility of agents is immovable and the multi-issue is redeemable-the utility loss of one issue can be recovered from the increased benefit of other issues for agents. But the actual conditions are not often like this. This paper proposed a multi-issue negotiation agents based on the forgetting theory. This method combined with logic and game theory can achieve the balance between negotiation logic rationality and utility rationality and dynamically adjust utility and deal with irredeemable problems. It is a new solution for multi-issue negotiation

关 键 词：多AGENT系统 多问题协商 遗忘理论 决策效用 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]