Agent信任与名誉的FIRE模型的改进

Improvement of FIRE:An Integrated Trust and Reputation Model for Open Multi-Agent Systems

在开放的多Agent系统中,信任与名誉对Agent之间的交互至关重要。FIRE是一种新型的、分布式的、采用多来源的综合性的信任与名誉模型。FIRE模型包括四个组成部分:IT(interaction trust)、RT(role-based trust)、WR(witness reputation)和CR(certified reputation),在与相关模型的比较中,显示出优越性。针对WR信息获取的低效、CR不能消除Agent的欺骗等不足,提出一种基于Agent组织的半分布式FIRE模型,相关分析和实验表明了改进的有效性。

Trust and reputation are central to effective interactions in OMAS （open multi-Agent system）. FIRE is a new, decentralized, and integrated trust and reputation model for OMAS, which consists of four parts： IT （interaction trust）, RT （role-based trust）, WR （witness reputation） and CR （certified reputation）. And it has shown advantages over other models. However, there are some drawbacks in FIRE such as the low efficiency in getting WR and the difficulty in eliminating possible lying from Agents. To solve the drawbacks, an improved organization-based and half-decentralized FIRE model was proposed. The results of contrastive experiment show that this improvement is effective.