Stereotypical trust and bias in dynamic multiagent systems

Abstract

Large-scale multiagent systems have the potential to be highly dynamic. Trust and reputation are crucial concepts in these environments, as it may be necessary for agents to rely on their peers to perform as expected, and learn to avoid untrustworthy partners. However, aspects of highly dynamic systems introduce issues which make the formation of trust relationships difficult. For example, they may be short-lived, precluding agents from gaining the necessary experiences to make an accurate trust evaluation. This article describes a new approach, inspired by theories of human organizational behavior, whereby agents generalize their experiences with previously encountered partners as stereotypes, based on the observable features of those partners and their behaviors. Subsequently, these stereotypes are applied when evaluating new and unknown partners. Furthermore, these stereotypical opinions can be communicated within the society, resulting in the notion of stereotypical reputation. We show how this approach can complement existing state-of-the-art trust models, and enhance the confidence in the evaluations that can be made about trustees when direct and reputational information is lacking or limited. Furthermore, we show how a stereotyping approach can help agents detect unwanted biases in the reputational opinions they receive from others in the society.

The Role of Trust in Human-Robot Interaction

“…However, whereby human judgements are often clouded by cultural or societal biases, stereotypes in MAS can be constructed in a way that maximizes the accuracy. Further work by the researchers in [14] shows how stereotypes in MAS can be spread throughout the group to improve others' trust assessments, and can be used by agents to detect unwanted biases received from others in the group. In [15], the authors show how this work can be used by organizations to create decision models based on trust assessments from stereotypes and other historical information about the other agents.…”

Stereotype Reputation with Limited Observability

“…In highly dynamic environments, where agents leave or depart regularly, relevant experience with trustees is often insufficient to produce reliable assessments. In these cases, stereotypes can be used to bootstrap trust and reputation [BNS13,LDR13,ŞY16].…”

“…Trust is the degree of belief, from the perspective of a trustor agent, that a trustee agent will act as they say they will in a given context [AG07,JIB07,BNS13]. A trustor with a high level of trust in a trustee is confident of a successful interaction with a good outcome.…”

“…In many domains, trustee agents exhibit traits that provide insight into their behaviour during, but can be observed prior to entering into, an interaction [BNS13,LDR13,ŞY16]. Such traits are referred to as stereotypes, and can be used to bootstrap trust and reputation assessments when experience is limited.…”

Trust beyond reputation: A computational trust model based on stereotypes

“…Honest and dishonest agents are biased towards different categories. A honest agent with probability 0.7 writes a review for a product from categories 1, 2, 3, 4; with probability 0.21 for products from categories 9, 10, 11, 12; and with probability 0.03 for products from categories 5,6,7,8. A dishonest agent with probability 0.7 writes a review for products from categories 5, 6, 7, 8; with probability 0.21 for products from categories 9, 10, 11, 12; and with probability 0.03 for products from categories 1, 2, 3, 4.…”