We consider the situation in which individuals in a finite population must repeatedly choose an action yielding an uncertain payoff. Between choices, each individual may observe the performance of one other individual. We search for rules of behavior with limited memory that increase expected...

Individuals repeatedly face a multi-decision task with unknown payoff distributions. They have minimal memory and update their strategy by observing previous play (and not strategy) of someone else. We select behavior rules that increase average payoffs as often as possible in a large population...

We analyze the evolution of behavioral rules for learning how to play a two-armed bandit. Individuals have no information about the underlying pay-off distributions and have limited memory about their own past experience. Instead they must rely on information obtained trough observing the...

The evolutionary selection of outcomes (modelled using the replicator dynamics) in games with costless communication depends crucially on the structural assumptions made on the underlying population. (1) In conflicts between two interacting populations, common interest implies that the set of...

The effect that exogenous mistakes, made by players choosing their strategies, have on the dynamic stability for the replicator dynamic is analyzed for both asymmetric and symmetric normal form games. Through these perturbed games, the dynamic solution concept of limit asymptotic stability is...

There are only few "positive" results concerning multi-person games with the fictitious play property, that is, games in which every fictitious play process approaches the set of equilibria. In this paper we characterize classes of multi-person games with the fictitious play property. We...

The analysis of the replicator dynamic in generic perfect information games yields the following results. In the long run, players play a Nash equilibrium provided that initially all strategies are present. There is at most one ``stable'' component (formally, an interior asymptotically stable...

A model of a population with a Local Interaction structure is presented. Individuals interact with others in a given Inter- action neighborhood to obtain their payof. Individuals either imitate or else they die and are replaced by one of their neigh- bors in another neighborhood- the Propagation...

In consectutive rounds, each agent in a finite population chooses an action, is randomly matched, obtains a payoff and then observes the performance of another agent. An agent determines future behavior based on the information she receives from the present round. She chooses among the...