A policy iteration algorithm for zero-sum stochastic games with mean payoff
نویسندگان
چکیده
منابع مشابه
Policy iteration algorithm for zero-sum multichain stochastic games with mean payoff and perfect information
We consider zero-sum stochastic games with finite state and action spaces, perfect information, mean payoff criteria, without any irreducibility assumption on the Markov chains associated to strategies (multichain games). The value of such a game can be characterized by a system of nonlinear equations, involving the mean payoff vector and an auxiliary vector (relative value or bias). We develop...
متن کاملA Potential Reduction Algorithm for Two-Person Zero-Sum Mean Payoff Stochastic Games
We suggest a new algorithm for two-person zero-sum undiscounted stochastic games focusing on stationary strategies. Given a positive real ǫ, let us call a stochastic game ǫ-ergodic, if its values from any two initial positions differ by at most ǫ. The proposed new algorithm outputs for every ǫ > 0 in finite time either a pair of stationary strategies for the two players guaranteeing that the va...
متن کاملTwo-Person Zero-Sum Stochastic Games with Semicontinuous Payoff
Consider a two-person zero-sum stochastic game with Borel state space S, compact metric action sets A, B and law of motion q such that the integral under q of every bounded Borel measurable function depends measurably on the initial state s and continuously on the actions (a,b) of the players. Suppose the payoff is a bounded function f of the infinite history of states and actions such that f i...
متن کاملA potential reduction algorithm for two-person zero-sum limiting average payoff stochastic games
We suggest a new algorithm for two-person zero-sum undiscounted stochastic games focusing on stationary strategies. Given a positive real , let us call a stochastic game -ergodic, if its values from any two initial points differ by at most . The proposed new algorithm outputs for every > 0 in finite time either a pair of stationary strategies for the two players guaranteeing that the values fro...
متن کاملStochastic Games with Parity Mean-payoff Objective
In this paper, we compute value of two-player games with perfect information equipped with the Par∧Avg>0 objectives. Moreover we show that even though the optimal strategies may require infinite memory, there exists an NP algorithm that computes the almost-sure region.
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Comptes Rendus Mathematique
سال: 2006
ISSN: 1631-073X
DOI: 10.1016/j.crma.2006.07.011