Practical Open-Loop Optimistic Planning - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Practical Open-Loop Optimistic Planning

Résumé

We consider the problem of online planning in a Markov Decision Process when given only access to a generative model, restricted to open-loop policies-i.e. sequences of actions-and under budget constraint. In this setting, the Open-Loop Optimistic Planning (OLOP) algorithm enjoys good theoretical guarantees but is overly conservative in practice, as we show in numerical experiments. We propose a modified version of the algorithm with tighter upper-confidence bounds, KL-OLOP, that leads to better practical performances while retaining the sample complexity bound. Finally, we propose an efficient implementation that significantly improves the time complexity of both algorithms.
Fichier principal
Vignette du fichier
407.pdf (932.66 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02375697 , version 1 (22-11-2019)

Identifiants

Citer

Edouard Leurent, Odalric-Ambrym Maillard. Practical Open-Loop Optimistic Planning. European Conference on Machine Learning, Sep 2019, Würzburg, Germany. ⟨hal-02375697⟩
42 Consultations
112 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More