Based On Online Learning Agent Negotiation

With the rapid development of Internet,the conventional means of operating business has greatly been affected.Currently,Agent-based e-commerce services has becoming a hotspot.How to solve the Agent negotiation learning problem quickly and high-efficently has been a mainstream in both economics and computer science domain.Agent based on negotiation is a decision-making process in incomplete information envionment among individual Agents whose behaviors are competitives, as well as cooperative.Since MAS is an open and dynamic system,negotiation process should adapt the change of such dynamic environment.Theories analysis show that,if learning mechanism can be embedded into multi-Agent based negotiation,which makes every Agent adjust its behaviors by learning,they will achieve the negotiation aims effectively.My paper is just based on the points mentioned above for further expansion in many related fields.The main topic of my paper is the research on how to use online learning to improve the negotiation efficiency in bilateral multi-issue negotiation.The effective negotiation model of the multi-Agent system are described in this paper, where the learning model of the agent negotiation of Zeng and Sycara(1997) are referenced.We embed Bayes learning mechanism on the basis of the negotiation model,and elaborate process descriptions of evaluating offers,belief revision and proposing counter-offers are presented.When both negotiators decide whether or not compromising so as to negotiation may be continue in stalemate in time-limited multi-issues among multi-Agents,negotiation strategies based on un-compromising degree is presented on the basis of Bayes learning,accordingly negotiation process does not end in some inessential conflicting point.The experimental results show that negotiation based on online learning should imporve predictive precision of opponent's private information, sequentially shorten negotiation time and advance negotiation efficiency.