Maximizing Expected Impact in an Agent Reputation Network

Rens, Gavin; Nayak, Abhaya C.; Meyer, Thomas

doi:10.1007/978-3-030-00111-7_9

Cited by 1 publication

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The RepNet-MDP framework [11] simplifies the notion of model by focusing in on key concepts such as behavioral habits and reputation of other agents. While this reduces the insights RepNet agents can have into other agents' behavior, it makes the framework arguably more intuitive.…”

Section: Related Mdp-based Frameworkmentioning

confidence: 99%

“…current value v, of the image on agent has of another agent, to be updated, and a new expected total impact i, of which the definition will be given shortly, U(v, i) returns an updated value of the image v . Many instantiations of this function are possible, two of which are presented in [11]. We will use the following instantiation:…”

Section: Formal Definition Of Repnet-mdpsmentioning

confidence: 99%

“…In 2018, Rens et al [11] proposed a mathematical framework, called RepNet-POMDP, designed to handle partially observable environments in which an agent's reputation among other agents dictates its behavior. The framework was subject to several mathematical inconsistencies, had no working implementation, and had a highly intractable planning algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Reputation-driven Decision-making in Networks of Stochastic Agents

Maoujoud¹,

Rens²

2020

Preprint

Self Cite

View full text Add to dashboard Cite

This paper studies multi-agent systems that involve networks of self-interested agents. We propose a Markov Decision Process-derived framework, called RepNet-MDP, tailored to domains in which agent reputation is a key driver of the interactions between agents. The fundamentals are based on the principles of RepNet-POMDP, a framework developed by Rens et al. [11] in 2018, but addresses its mathematical inconsistencies and alleviates its intractability by only considering fully observable environments. We furthermore use an online learning algorithm for finding approximate solutions to RepNet-MDPs. In a series of experiments, RepNet agents are shown to be able to adapt their own behavior to the past behavior and reliability of the remaining agents of the network. Finally, our work identifies a limitation of the framework in its current formulation that prevents its agents from learning in circumstances in which they are not a primary actor.

show abstract

Section: Related Mdp-based Frameworkmentioning

confidence: 99%

Section: Formal Definition Of Repnet-mdpsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation