Learning to Play the Chess Variant Crazyhouse Above World Champion Level With Deep Neural Networks and Human Data.

Monte-Carlo tree search chess crazyhouse deep learning supervised learning

Journal

Frontiers in artificial intelligence
ISSN: 2624-8212
Titre abrégé: Front Artif Intell
Pays: Switzerland
ID NLM: 101770551

Informations de publication

Date de publication:
2020
Historique:
received: 20 08 2019
accepted: 26 03 2020
entrez: 18 3 2021
pubmed: 19 3 2021
medline: 19 3 2021
Statut: epublish

Résumé

Deep neural networks have been successfully applied in learning the board games Go, chess, and shogi without prior knowledge by making use of reinforcement learning. Although starting from zero knowledge has been shown to yield impressive results, it is associated with high computationally costs especially for complex games. With this paper, we present

Identifiants

pubmed: 33733143
doi: 10.3389/frai.2020.00024
pmc: PMC7861260
doi:

Types de publication

Journal Article

Langues

eng

Pagination

24

Informations de copyright

Copyright © 2020 Czech, Willig, Beyer, Kersting and Fürnkranz.

Références

Nature. 2016 Jan 28;529(7587):484-9
pubmed: 26819042
Nature. 2017 Oct 18;550(7676):354-359
pubmed: 29052630
Science. 2018 Dec 7;362(6419):1140-1144
pubmed: 30523106

Auteurs

Johannes Czech (J)

Department of Computer Science, TU Darmstadt, Darmstadt, Germany.

Moritz Willig (M)

Department of Computer Science, TU Darmstadt, Darmstadt, Germany.

Alena Beyer (A)

Department of Computer Science, TU Darmstadt, Darmstadt, Germany.

Kristian Kersting (K)

Department of Computer Science, TU Darmstadt, Darmstadt, Germany.
Centre for Cognitive Science, TU Darmstadt, Darmstadt, Germany.

Johannes Fürnkranz (J)

Department of Computer Science, JKU Linz, Linz, Austria.

Classifications MeSH