Some results on approximating the minimax solution in approval voting

Voting has been a very popular method for preference aggregation in multiagent environments. It is often the case that a set of agents with different preferences need to make a choice among a set of alternatives, where the alternatives could be various entities such as potential committee members, or joint plans of action. A standard methodology for this scenario is to have each agent express his preferences and then select an alternative according to some voting protocol. Several decision making applications in AI have followed this approach including problems in collaborative filtering [10] and planning [3, 4].