 |
|
|
|
| HAL: hal-00403463, version 4 |
| Detailed view |  Export this paper |
|
|
| Available versions: | v1 (2009-07-10) | v2 (2009-07-11) | v3 (2009-10-09) | v4 (2009-12-03) |
|
|
|
|
| Randomized Strategies are Useless in Markov Decision Processes |
|
|
| Hugo Gimbert 1 |
|
|
| (2009-07-10) |
|
|
|
| We show that in a Markov decision process with arbitrary payoff mapping, restricting the set of behavioral strategies from randomized to deterministic does not influence the value of the game nor the existence of almost-surely or positively winning strategies. This result still holds for Markov decision processes with partial observation. |
|
|
|
|
|
|
|
|
|
|
|
| 1: | Laboratoire Bordelais de Recherche en Informatique (LaBRI) |
|
|
|
CNRS : UMR5800 – Université Sciences et Technologies - Bordeaux I – Ecole Nationale Supérieure d'Electronique, Informatique et Radiocommunications de Bordeaux – Université Victor Segalen - Bordeaux II |
|
|
|
|
|
|
|
|
|
| Méthodes Formelles, LaBRI |
|
|
|
|
| Subject | : | Computer Science/Computer Science and Game Theory |
|
|
| Attached file list to this document: | |||||
|
|||||
|
|
|
| hal-00403463, version 4 | |
| http://hal.archives-ouvertes.fr/hal-00403463 | |
| oai:hal.archives-ouvertes.fr:hal-00403463 | |
| From: Hugo Gimbert | |
| Submitted on: Thursday, 3 December 2009 14:31:56 | |
| Updated on: Thursday, 3 December 2009 14:44:14 | |
