Optimal policies for constrained average-cost Markov decision processes
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.
| Autores principales: | , |
|---|---|
| Formato: | Artículo |
| Lenguaje: | inglés |
| Publicado: |
2009
|
| Materias: | |
| Acceso en línea: | http://eprints.uanl.mx/1957/1/Top.pdf |
| Sumario: | 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. |
|---|