Exploiting separability in multiagent planning with continuous-state mdps (extended abstract)

Jilles Steeve Dibangoye 1, 2 Christopher Amato 3 Olivier Buffet 4 François Charpillet 5
1 CHROMA - Robots coopératifs et adaptés à la présence humaine en environnements dynamiques
CITI - CITI Centre of Innovation in Telecommunications and Integration of services, Inria Grenoble - Rhône-Alpes
4 MAIA - Autonomous intelligent machine
Inria Nancy - Grand Est, LORIA - AIS - Department of Complex Systems, Artificial Intelligence & Robotics
5 LARSEN - Lifelong Autonomy and interaction skills for Robots in a Sensing ENvironment
Inria Nancy - Grand Est, LORIA - AIS - Department of Complex Systems, Artificial Intelligence & Robotics
Abstract : Decentralized partially observable Markov deci- sion processes (Dec-POMDPs) provide a general model for decision-making under uncertainty in co- operative decentralized settings, but are difficult to solve optimally (NEXP-Complete). As a new way of solving these problems, we recently intro- duced a method for transforming a Dec-POMDP into a continuous-state deterministic MDP with a piecewise-linear and convex value function. This new Dec-POMDP formulation, which we call an occupancy MDP, allows powerful POMDP and continuous-state MDP methods to be used for the first time. However, scalability remains limited when the number of agents or problem variables becomes large. In this paper, we show that, un- der certain separability conditions of the optimal value function, the scalability of this approach can increase considerably. This separability is present when there is locality of interaction be- tween agents, which can be exploited to improve performance. Unlike most previous methods, the novel continuous-state MDP algorithm retains op- timality and convergence guarantees. Results show that the extension using separability can scale to a large number of agents and domain variables while maintaining optimality.
Type de document :
Communication dans un congrès
Proceedings of the twenty-fourth international joint conference on artificial intelligence, IJCAI 2015, Jul 2015, Buenos Aires, Argentina. 2015
Liste complète des métadonnées

https://hal.inria.fr/hal-01188483
Contributeur : Jilles Steeve Dibangoye <>
Soumis le : lundi 31 août 2015 - 00:08:11
Dernière modification le : mercredi 6 décembre 2017 - 10:44:09

Identifiants

  • HAL Id : hal-01188483, version 1

Citation

Jilles Steeve Dibangoye, Christopher Amato, Olivier Buffet, François Charpillet. Exploiting separability in multiagent planning with continuous-state mdps (extended abstract). Proceedings of the twenty-fourth international joint conference on artificial intelligence, IJCAI 2015, Jul 2015, Buenos Aires, Argentina. 2015. 〈hal-01188483〉

Partager

Métriques

Consultations de la notice

141