A Two-Step Minimax Q-learning Algorithm for Two-Player Zero-Sum Markov Games

An interesting iterative procedure is proposed to solve a two-player zero-sum Markov games. First this problem is expressed as a min-max Markov game. Next, a two-step Q-learning algorithm for solving Markov decision problem (MDP) is suitably modified to solve this Markov game. Under a suitable assumption, the boundedness of the proposed iterates is obtained theoretically. Using results from stochastic approximation, the almost sure convergence of the proposed two-step minimax Q-learning is obtained theoretically. More specifically, the proposed algorithm converges to the game theoretic optimal value with probability one, when the model information is not known. Numerical simulation authenticate that the proposed algorithm is effective and easy to implement.