NO FREE LUNCHES IN MULTI-AGENT SYSTEMS
Stefan J. Johansson · International Journal of Computational Intelligence and Applications · 2002
We introduce the notion of Characteristic Distributions which is a way of representing information about the payoffs of different behaviors in a Multi-Agent System. We discuss how they can be used to simplify and structure the analysis of strategies and prove the existence of optimal environments, given a certain behavior, and that all behaviors payoff equally, when taken over all possible environments (no free lunch theorem for strategies).