Error-Bounded Approximations for Infinite-Horizon Discounted Decentralized POMDPs

Jilles Steeve Dibangoye 1 Olivier Buffet 1 François Charpillet 1
1 MAIA - Autonomous intelligent machine
Inria Nancy - Grand Est, LORIA - AIS - Department of Complex Systems, Artificial Intelligence & Robotics
Abstract : We address decentralized stochastic control problems represented as decentralized partially observable Markov decision processes (Dec-POMDPs). This formalism provides a general model for decision-making under uncertainty in cooperative, decentralized settings, but the worst-case complexity makes it difficult to solve optimally (NEXP-complete). Recent advances suggest recasting Dec-POMDPs into continuous-state and deterministic MDPs. In this form, however, states and actions are embedded into high-dimensional spaces, making accurate estimate of states and greedy selection of actions intractable for all but trivial-sized problems. The primary contribution of this paper is the first framework for error-monitoring during approximate estimation of states and selection of actions. Such a framework permits us to convert state-of-the-art exact methods into error-bounded algorithms, which results in a scalability increase as demonstrated by experiments over problems of unprecedented sizes.
Type de document :
Communication dans un congrès
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD), Sep 2014, Nancy, France. 8724, pp.338 - 353, 2014, Machine Learning and Knowledge Discovery in Databases. 〈http://www.ecmlpkdd2014.org/〉. 〈10.1007/978-3-662-44848-9_22〉
Liste complète des métadonnées

Littérature citée [21 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01096610
Contributeur : Olivier Buffet <>
Soumis le : mercredi 17 décembre 2014 - 17:32:39
Dernière modification le : jeudi 11 janvier 2018 - 06:25:23
Document(s) archivé(s) le : lundi 23 mars 2015 - 16:05:35

Fichier

ecml14.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Jilles Steeve Dibangoye, Olivier Buffet, François Charpillet. Error-Bounded Approximations for Infinite-Horizon Discounted Decentralized POMDPs. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD), Sep 2014, Nancy, France. 8724, pp.338 - 353, 2014, Machine Learning and Knowledge Discovery in Databases. 〈http://www.ecmlpkdd2014.org/〉. 〈10.1007/978-3-662-44848-9_22〉. 〈hal-01096610〉

Partager

Métriques

Consultations de la notice

212

Téléchargements de fichiers

279