Resumen de Optimal policies for constrained average-cost Markov decision processes

Juan González Hernández, César E. Villarreal Villarreal

• We give mild conditions for the existence of optimal solutions for a Markov decision problem with average cost, under m constraints of the same kind, in Borel actions and states spaces. Moreover, there is an optimal policy that is a convex combination of at most m+1 deterministic policies.