Improving trust and reputation assessment with dynamic behaviour

Abstract

Trust between agents in multi-agent systems (MASs) is critical to encourage high levels of cooperation. Existing methods to assess trust and reputation use direct and indirect past experiences about an agent to estimate their future performance; however, these will not always be representative if agents change their behaviour over time. Real-world distributed networks such as online market places, P2P networks, pervasive computing and the Smart Grid can be viewed as MAS. Dynamic agent behaviour in such MAS can arise from seasonal changes, cheaters, supply chain faults, network traffic and many other reasons. However, existing trust and reputation models use limited techniques, such as forgetting factors and sliding windows, to account for dynamic behaviour. In this paper, we propose Reacting and Predicting in Trust and Reputation (RaPTaR), a method to extend existing trust and reputation models to give agents the ability to monitor the output of interactions with a group of agents over time to identify any likely changes in behaviour and adapt accordingly. Additionally, RaPTaR can provide an a priori estimate of trust when there is little or no interaction data (either because an agent is new or because a detected behaviour change suggests recent past experiences are no longer representative). Our results show that RaPTaR has improved performance compared to existing trust and reputation methods when dynamic behaviour causes the ranking of the best agents to interact with to change.

Using Change Detection to Adapt to Dynamically Changing Trustees

“…We compared BRS-PH to BRS (BRS-Basic) [1], BRS with a forgetting factor (BRS-FF) [1], RaPTaR [12], as well as an HMM-based trust model (HMM) [4]. We ran each of the methods with every setting listed in Section 4.1.…”

“…Player and Griffiths proposed RaPTaR [12], a method to extend existing trust models to detect and adjust to behavior changes. RaPTaR has a learning component and a predictive component.…”

“…The PH test is a well-known change detector in signal processing [10,11]. To the best of our knowledge, the only existing method using suchlike statistical test for detcting dynamic trustees is RaPTaR [12], which uses the Kolmogorov-Smirnov test instead of PH.…”

“…Especially noteworthy is the simplicity of our system. It significantly outperforms an HMM method [4] and RaPTaR [12], while at the same time being much simpler in design and implementation. This is important because it demonstrates that a complex system is not necessary to achieve state-of-the-art performance in handling dynamic trustees.…”

Special issue on adaptive and learning agents 2018

“…In the paper Improving trust and reputation assessment with dynamic behaviour, Player and Griffiths (2020) propose Reacting and Predicting in Trust and Reputation (RaPTaR), a method that extends existing trust and reputation models to give agents the ability to monitor the output of interactions with a group of agents over time to identify any likely changes in behaviour and adapt accordingly. Experiments were conducted where agents selected partners for tasks using small-world networks, scale-free networks and fully connected networks, and a number of previously published trust and reputation models were evaluated alongside the proposed RaPTaR method.…”