This article investigates the conjecture that agents who make decisions in scenarios where trust is important can benefit from the use of asocial structure, representing the social relationships that exist between agents. We propose techniques that can be used by agents to initially build and then progressively update such a structure in the light of experience. We describe an implementation of our techniques in the domain of the Agent ART testbed: we take two existing agents for this domain (“Simplet” and “Connected”) and compare their performance with versions that use our social structure (“SocialSimplet” and “SocialConnected”). We show that SocialSimplet and SocialConnected outperform their counterparts with respect to the quality of the interactions, the number of rounds won in a competition, and the total utility gained.
No takes yet. Share an insight, caveat, or question.
Erriquez et al. (2013) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: