Solving POMDPs using selected past events - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2000

Solving POMDPs using selected past events

Alain Dutech

Résumé

A new algorithm for solving Partially Observed Markov Decision Processes build on theoretical results about exhaustive observable. It aims at building an extension of the state space so as to obtain a Markov Decision Process which can then be solved by classical methods. We present two versions of the algorithm, one using reinforcement learning when the evolution model is unknown, the other is quicker but requires the knowledge of this evolution model.

Domaines

Autre [cs.OH]
Fichier non déposé

Dates et versions

inria-00099378 , version 1 (26-09-2006)

Identifiants

  • HAL Id : inria-00099378 , version 1

Citer

Alain Dutech. Solving POMDPs using selected past events. European Conference on Artificial Intelligence, 2000, Berlin, Germany. ⟨inria-00099378⟩
82 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More