Reference : Monte-Carlo Tree Search in Backgammon
Scientific congresses and symposiums : Paper published in a book
Engineering, computing & technology : Computer science
http://hdl.handle.net/2268/28469
Monte-Carlo Tree Search in Backgammon
English
Van Lishout, François mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
Chaslot, Guillaume mailto [Universiteit Maastricht > > > >]
Uiterwijk, Jos W.H.M. mailto [Universiteit Maastricht > > > >]
2007
175-184
Yes
Computer Games Workshop
June 15-17, 2007
IBM, SARA, NCF, ICGA, MICC-IKAT
Amsterdam
The Netherlands
[en] Backgammon ; Monte-Carlo Tree Search
[en] Monte-Carlo Tree Search is a new method which has been applied successfully to many games. However, it has never been tested on two-player perfect-information games with a chance factor. Backgam- mon is the reference game of this category. Today’s best Backgammon programs are based on reinforcement learning and are stronger than the best human players. These programs have played millions of offline games to learn to evaluate a position. Our approach consists rather in playing online simulated games to learn how to play correctly in the current position.
http://hdl.handle.net/2268/28469

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
vanlishout_backgammon.pdfAuthor postprint250.98 kBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.