Distributed Methods and Processes in Games of Incomplete Information

摘要

One factor limiting the applicability of game theory has been the inadequacy ofcomputational tools for handling real world problems where decision makers have incomplete information. The thesis presents the first approach to computing Pareto optimal game solutions under such circumstances, and thus opens new research possibilities. Moreover, the reachability of Pareto optimal agreements in negotiations is studied under realistic assumptions about information. The methods developed in the thesis may find applications in problems where independent decision makers, each having limited information about the situation, need to cooperate or to reach a joint decision. Another main topic of the thesis is the illumination of the interdisciplinary nature of this problem type. The third topic is the stability of asynchronous iterations that are relevant in game-related reaction processes and in distributed computation. The thesis consists of a summary and the six papers included in this volume.