摘要
研究Agent多次协商过程中的策略调整问题,目的是使得Agent在协商过程中具有自学能力,对环境和协商对手更敏感。结合资源分配问题,讨论Agent协商过程中的学习问题,基于博弈论分别分析了单次协商和多次协商模型,给出了协商过程中在不同信息条件下遵循的策略,并进行了证明。
For the purpose of making agent to be more intelligent to its environments and opponents, the learning and coordination in the process of multiple negotiationes concerning the problem of RA are discussed. Based on conclusions from game theory, the single negotiation and multiple negotiation model are analyzed. In each of the two models, the strategies as well as its proof are given that the agents should follow under different information conditions.
出处
《计算机工程与设计》
CSCD
北大核心
2005年第6期1612-1613,1644,共3页
Computer Engineering and Design
