Bayesian Learning, Smooth Approximate Optimal Behavior, and Convergence toε-Nash Equilibrium
Econometrica · 2015 · Volume 83 · Issue 1
Abstract
In this paper, I construct players' prior beliefs and show that these prior beliefs lead the players to learn to play an approximate Nash equilibriumuniformlyin any infinitely repeated slightly perturbed game with discounting and perfect monitoring. That is, given anyε> 0, there exists a (single) profile of players' prior beliefs that leads play to almost surely converge to anε-Nash equilibrium uniformly for any (finite normal form) stage game with slight payoff perturbation and any discount factor less than 1.
Keywords
Bayesian learning, smooth approximate optimal behavior, convergence, ε-Nash equilibrium, η-learning to predict strategies
Enable JavaScript for citation tools and additional paper interactions.