Modification of UCT with Patterns in Monte-Carlo Go - Archive ouverte HAL Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 2006

Modification of UCT with Patterns in Monte-Carlo Go

Résumé

Algorithm UCB1 for multi-armed bandit problem has already been extended to Algorithm UCT (Upper bound Confidence for Tree) which works for minimax tree search. We have developed a Monte-Carlo Go program, MoGo, which is the first computer Go program using UCT. We explain our modification of UCT for Go application and also the intelligent random simulation with patterns which has improved significantly the performance of MoGo. UCT combined with pruning techniques for large Go board is discussed, as well as parallelization of UCT. MoGo is now a top level Go program on $9\times9$ and $13\times13$ Go boards.
Fichier principal
Vignette du fichier
RR-6062.pdf (630.71 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

inria-00117266 , version 1 (30-11-2006)
inria-00117266 , version 2 (12-12-2006)
inria-00117266 , version 3 (20-12-2006)

Identifiants

  • HAL Id : inria-00117266 , version 3

Citer

Sylvain Gelly, Yizao Wang, Rémi Munos, Olivier Teytaud. Modification of UCT with Patterns in Monte-Carlo Go. [Research Report] RR-6062, INRIA. 2006. ⟨inria-00117266v3⟩
2538 Consultations
9896 Téléchargements

Partager

Gmail Facebook X LinkedIn More