Abstract: We first review the basic models of one-shot and repeated games of incomplete information, and the corresponding concept of Bayesian-Nash equilibrium. Guided by expected utility maximization, a rational learning process leads the players from Bayesian equilibrium of the incomplete information repeated game to Nash equilibrium of the complete information repeated game. Exploration-exploitation tradeoffs are determined endogenously and dynamically as the game progresses.

Some references:
Fudenberg D and D.K. Levine, "The Theory of Learning in Games," MIT Press, 1998
Jordan, J. S. "Bayesian Learning in Normal Form Games," Games and Economic Behavior, 3, 60-81, 1992
Jackson, M.O. E. Kalai and R. Smorodinsky, "Bayesian Representation of Stochastic Processes under Learning: de Finetti Revisited," Econometrica, 67, No. 4, 875-794 1999
Kalai, E. and E. Lehrer, "Rational Learning Leads to Nash Equilibrium," Econometrica Vol. 61, No. 5, 1993, pp. 1019-1045. 1993
Sorin, Sylvain, "Merging, Reputation, and Repeated Games with Incomplete Information," Games and Economic Behavior, vol. 29(1-2), pages 274-308, 1999.