Learning opening books in partially observable games: using random seeds in Phantom Go

Abstract : Many artificial intelligences (AIs) are randomized. One can be lucky or unlucky with the random seed; we quantify this effect and show that, maybe contrarily to intuition, this is far from being negligible. Then, we apply two different existing algorithms for selecting good seeds and good probability distributions over seeds. This mainly leads to learning an opening book. We apply this to Phantom Go, which, as all phantom games, is hard for opening book learning. We improve the winning rate from 50% to 70% in 5x5 against the same AI, and from approximately 0% to 40% in 5x5, 7x7 and 9x9 against a stronger (learning) opponent.
Type de document :
Communication dans un congrès
Computer intelligence and Games (CIG 2016), Sep 2016, Santorini, Greece. 2016, Computer intelligence and Games
Liste complète des métadonnées

https://hal.inria.fr/hal-01413229
Contributeur : Fabien Teytaud <>
Soumis le : vendredi 9 décembre 2016 - 15:08:26
Dernière modification le : jeudi 5 avril 2018 - 12:30:12
Document(s) archivé(s) le : jeudi 23 mars 2017 - 10:22:09

Fichier

seedPhantomGo.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01413229, version 1

Collections

Citation

Tristan Cazenave, Jialin Liu, Fabien Teytaud, Olivier Teytaud. Learning opening books in partially observable games: using random seeds in Phantom Go. Computer intelligence and Games (CIG 2016), Sep 2016, Santorini, Greece. 2016, Computer intelligence and Games. 〈hal-01413229〉

Partager

Métriques

Consultations de la notice

353

Téléchargements de fichiers

60