Report on Satisfaction Equilibria ?

In real life problems, agents are generally faced with situations where they only have partial or no knowledge about their environment and the other agents evolving in it. So far, most game theory approaches have assumed that this knowledge about the other agents was known by all agents and that observations could be made about all the agents’ actions. In order to alleviate these assumptions, we propose a new game model in which all an agent knows is its own actions and rewards for certain specific outcomes it observes. We show that in this type of games, all an agent can do is reasoning about its own payoffs and that therefore, classical equilibria through deliberation are not possible. To palliate to this difficulty, we introduce the satisfaction principle from which an equilibrium can arise as the result of the agents individual learning experiences. We define such an equilibrium and then we present different algorithms that can be used to reach it. Finally, we present experimental results and theoretical proofs that show that using learning strategies based on this specific equilibrium, agents will generally coordinate themselves on a Pareto-optimal joint strategy, that is not always a Nash equilibrium, even though each agent is individually rational, in the sense that they try to maximize their own satisfaction.