Social Interaction Under Uncertainty in Multi Agent Systems

Abstract

Multi-agent systems deal with environments comprising several agents that interact with each other. The development of distributed, interconnected computer systems has invoked the rapid growth of this research area. Such settings, where one software agent interacts with another, require studying interactions such as coordination, cooperation and negotiation. In fact, many types of social interactions among humans relate to computerized agents as well. In my work I investigate computational aspects of two common social interactions, namely voting and collaborative search. I also propose to solve my suggested problems without many relaxing assumptions in order to come as close as possible to real-world settings. Hence, I concentrate on settings of imperfect information which I model using probabilities.