arXiv subjects
Publications and source records attributed to K. Avrachenkov.
We consider a finite state, finite action, zero-sum stochastic games with data defining the game lying in the ordered field of algebraic numbers. In both the discounted and the limiting average versions of these games we prove that the value vector also lies in the same field of algebraic numbers. In a prescribed sense, our results settle a problem that has remained open since, at least, 1991.
We consider a non-cooperative constrained stochastic games with N players with the following special structure. With each player there is an associated controlled Markov chain. The transition probabilities of the i-th Markov chain depend only on the state and actions of controller i. The information structure that we consider is such that each player knows the state of its own MDP and its own actions. It does not know the states of, and the actions taken by other players. Finally, each player wishes to minimize a time-average cost function, and has constraints over other time-avrage cost functions. Both the cost that is minimized as well as those defining the constraints depend on the state and actions of all players. We study in this paper the existence of a Nash equilirium. Examples in power control in wireless communications are given.