Learning a Move-Generator for Upper Con dence Trees - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2012

Learning a Move-Generator for Upper Con dence Trees

Abstract

We experiment the introduction of machine learning tools to improve Monte-Carlo Tree Search. More precisely, we propose the use of Direct Policy Search, a classical reinforcement learning paradigm, to learn the Monte-Carlo Move Generator. We experiment our algorithm on di erent forms of unit commitment problems, including experiments on a problem with both macrolevel and microlevel decisions.
Fichier principal
Vignette du fichier
uctWithDPS.pdf (220 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00759822 , version 1 (03-12-2012)

Identifiers

  • HAL Id : hal-00759822 , version 1

Cite

Adrien Couetoux, Olivier Teytaud, Hassen Doghmen. Learning a Move-Generator for Upper Con dence Trees. International Computer Symposium 2012, Dec 2012, Hualien, Taiwan. ⟨hal-00759822⟩
278 View
281 Download

Share

Gmail Facebook Twitter LinkedIn More