This paper proposes a method allowing an agent to perform in a socially fair way by considering other agents' preferences. Such a balanced action selection process is based on declarative diagnosis, which enables the removal of contradictions arising as all agents' preferences are confronted within the deciding agent. Agents can be negatively affected when some of their preferences are not respected for the sake of a global compromise. The set of preferences to be yielded by agents in order to remove all contradictions in a balanced way (i.e. the diagnosis that better manages how each agent is to be affected) is determined by minimising a cost function computed over application independent features. By respecting the resulting non-contradictory preferences set, the deciding agent acts cooperatively. © Springer-Verlag Berlin Heidelberg 2006.
CITATION STYLE
Santana, P., & Pereira, L. M. (2006). Emergence of cooperation through mutual preference revision. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4031 LNAI, pp. 81–90). Springer Verlag. https://doi.org/10.1007/11779568_11
Mendeley helps you to discover research relevant for your work.