We investigate the problem of cooperation of self-interested agents with respect to the confidentiality of their plans and without a presence of any third-party mediator. We base our approach on non-zero-sum n-player games in the extensive form focusing on two main goals: (1) the analysis of the maximal improvement of the utility value that an agent can gain by negotiation, and (2) the dependence of this improvement on the basic characteristics of the game - the number of agents, size of the branching factor, and correlation of the utility functions. We provide theoretical definitions together with experimental results evaluated on games abstracting real-world scenarios. © 2011 Springer-Verlag.
CITATION STYLE
Bošanský, B., Lisý, V., & Pěchouček, M. (2011). Towards cooperation in adversarial search with confidentiality. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6867 LNAI, pp. 246–255). https://doi.org/10.1007/978-3-642-23181-0_24
Mendeley helps you to discover research relevant for your work.