New Algorithms for Solving Zero-Sum Stochastic Games

Zero-sum stochastic games, henceforth are a classical model in game theory which two opponents interact and the environment changes response to players’ behavior. The central solution concepts for these games discounted values value, represent what playing is worth players different levels of impatience. In present manuscript, we provide algorithms computing exact expressions polynomial number pure stationary strategies players. This result considerably improves all existing algorithms.