On A Markov Game With One-sided Incomplete Information
Source: Yale University
The authors apply the average cost optimality equation to zero-sum Markov games, by considering a simple game with one-sided incomplete information that generalizes an example of Aumann and Maschler (1995). They determine the value and identify the optimal strategies for a range of parameters. Dynamic games with incomplete information have a long history in economics. With few exceptions, these games model the uncertainty as fixed throughout the game, and private signals about the state of nature are observed only once at the beginning of the game. In many of these models, information is asymmetric, with one player's information being finer than his opponent's.