Planification et apprentissage par renforcement avec modèles d'actions compacts

Boris Lesner 1
1 Equipe Hultech - Laboratoire GREYC - UMR6072
GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
Abstract : We study Markovian Decision Processes represented with Probabilistic STRIPS action models. A first part of our works is about solving those processes in a compact way. To that end we propose two algorithms. A first one based on propositional formula manipulation allows to obtain approximate solutions in tractable propositional fragments such as Horn and 2-CNF. The second algorithm solves exactly and efficiently problems represented in PPDDL using a new notion of extended value functions. The second part is about learning such action models. We propose different approaches to solve the problem of ambiguous observations occurring while learning. Firstly, a heuristic method based on Linear Programming gives good results in practice yet without theoretical guarantees. We next describe a learning algorithm in the “Knows What It Knows” framework. This approach gives strong theoretical guarantees on the quality of the learned models as well on the sample complexity. These two approaches are then put into a Reinforcement Learning setting to allow an empirical evaluation of their respective performances.
Document type :
Theses
Liste complète des métadonnées

Cited literature [73 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/tel-01076437
Contributor : Greyc Référent <>
Submitted on : Wednesday, October 22, 2014 - 10:58:16 AM
Last modification on : Sunday, April 21, 2019 - 10:33:55 AM
Document(s) archivé(s) le : Friday, January 23, 2015 - 10:30:30 AM

Identifiers

  • HAL Id : tel-01076437, version 1

Citation

Boris Lesner. Planification et apprentissage par renforcement avec modèles d'actions compacts. Apprentissage [cs.LG]. université de caen, 2011. Français. ⟨tel-01076437⟩

Share

Metrics

Record views

394

Files downloads

185