Towards a solution of 7x7 Go with Meta-MCTS

Abstract : Solving board games is a hard task, in particular for games in which classical tools like alpha-beta and proof-number-search are some- how weak. In particular, Go is not solved (in any sense of solving, even the weakest) beyond 6x6. We here investigate the use of Meta-Monte- Carlo-Tree-Search, for building a huge 7x7 opening book. In particular, we report the twenty wins (out of twenty games) that were obtained re- cently in 7x7 Go against pros; we also show that in one of the games, with no human error, the pro might have won.
Type de document :
Communication dans un congrès
Advances in Computer Games, 2011, Tilburg, Netherlands. International Conference on Computers and Games 2011, pp.12, 2012, International Conference on Computers and Games 2011
Liste complète des métadonnées

https://hal.inria.fr/hal-01077624
Contributeur : Olivier Teytaud <>
Soumis le : dimanche 26 octobre 2014 - 01:45:00
Dernière modification le : mercredi 28 novembre 2018 - 15:36:02

Identifiants

  • HAL Id : hal-01077624, version 1

Collections

Citation

Cheng-Wei Chou, Li-Wen Wu, Hui-Min Wang, Fabien Teytaud, Olivier Teytaud, et al.. Towards a solution of 7x7 Go with Meta-MCTS. Advances in Computer Games, 2011, Tilburg, Netherlands. International Conference on Computers and Games 2011, pp.12, 2012, International Conference on Computers and Games 2011. 〈hal-01077624〉

Partager

Métriques

Consultations de la notice

373