Constrained Markov Decision Chains

We consider finite state and action discrete time parameter Markov decision chains. The objective is to provide an algorithm for finding a policy that minimizes the long-run expected average cost when there are linear side conditions on the limit points of the expected state-action frequencies. This problem has been solved previously only for the case where every deterministic stationary policy has at most one ergodic class. This note removes that restriction by applying the Dantzig-Wolfe decomposition principle.

• 저널Management Science · 19(4-part-1) · 389–390
• 토픽Reinforcement Learning in Robotics · Artificial Intelligence
• DOI10.1287/mnsc.19.4.389
• 저자Cyrus Derman, Arthur F. Veinott