Learning in Bayesian Games by Bounded Rational Players I
نویسندگان
چکیده
We study learning in Bayesian games (or games with differential information) with an arbitrary number of bounded rational players, i.e., players who choose approximate best response strategies [approximate Bayesian Nash Equilibrium (BNE) strategies] and who also are allowed to be completely irrational in some states of the world. We show that bounded rational players by repetition can reach a limit full information BNE outcome. We also prove the converse, i.e., given a limit full information BNE outcome, we can construct a sequence of bounded rational plays that converges to the limit full information BNE outcome.
منابع مشابه
Learning in Bayesian Games by Bounded Rational Players Ii: Nonmyopia
We generalize results of earlier work on learning in Bayesian games by allowing players to make decisions in a nonmyopic fashion. In particular, we address the issue of nonmyopic Bayesian learning with an arbitrary number of bounded rational players, i.e., players who choose approximate best-response strategies for the entire horizon (rather than the current period). We show that, by repetition...
متن کاملFinite Stage Centipede Game and other Finite Two Player Dynamic Games with Bounded Rationality
Subgame Perfect Nash Equilibrium (SPNE), as a predictive theory, has been found to be inconsistent with experimental evidence in many dynamic games. This working paper introduces bounded rationality of varying degrees through a distribution of the types of each player in finite dynamic games, and checks if that rationalizes observed/“intuitive” behavior. Each of these types has a different degr...
متن کاملBoundedly Rational Quasi-bayesian Learning in Coordination Games with Imperfect Monitoring
In this paper we study players’ long-run behaviors in evolutionary coordination games with imperfect monitoring. In each time period, signals corresponding to players’ underlying actions, instead of the actions themselves, are observed. A boundedly rational quasi-Bayesian learning process is proposed to extract information from the realized signals. We find that players’ long-run behaviors depe...
متن کاملRobust Opponent Modeling in Real-Time Strategy Games using Bayesian Networks
Opponent modeling is a key challenge in Real-Time Strategy (RTS) games as the environment is adversarial in these games, and the player cannot predict the future actions of her opponent. Additionally, the environment is partially observable due to the fog of war. In this paper, we propose an opponent model which is robust to the observation noise existing due to the fog of war. In order to cope...
متن کاملLearning to Play Equilibria: The Bayesian, Repeated Games Approach∗
This is a survey of the last 15 years or so of research into an important class of models of how players might come to play equilibria in games: Bayesian models of learning in repeated strategic form games. There have been two main objectives for this research. One has been to tell stories for how equilibrium might arise, with the ultimate aim of developing a predictive theory. The other has be...
متن کامل