How Coalitions Enhance Cooperation in the IPD over Complex Networks

Grouping agents into coalitions can be an important way of cooperation through which multi-agent systems(MAS) improve their performance, accomplish their assignments, increase their benefits, and achieve their goals. However, agents belonging to a coalition must decide how they behave to increase their gains. In this paper, we present a model that combines learning and coalition formation to enhance cooperation over the Iterated Prisoner´s Dilemma (IPD). We focus on agents interacting over complex networks since they provide a realistic model of the nowadays interconnected world. Our results suggest that coalitions are a relevant contribution to achieve cooperation in the IPD.