Talk:Havannah (board game)
 | This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | ||||||||||
| |||||||||||
Computational intelligence: using a mix of
- zero-learning as in AlphaZero
- boardsize invariance thanks to fully convolutional neural networks (as in U-Net) and Convolutional_neural_network#Pooling
a team at https://fr.wikipedia.org/wiki/Facebook_Artificial_Intelligence_Research, National_Dong_Hwa_University, University_of_the_Littoral_Opal_Coast, Lamsade and others successfully implemented a bot playing super-human level at Havannah: http://littlegolem.net/jsp/info/player.jsp?plid=138065 The program notably played surprising openings, leading to very short wins.
The present contributor is part of that team, so I do not modify the page myself, I just discuss here. 2A01:E0A:544:29F0:4A8:44E4:851:F90B (talk) 08:10, 28 May 2020 (UTC) 2A01:E0A:544:29F0:4A8:44E4:851:F90B (talk) 08:38, 28 May 2020 (UTC)
- The list of authors is incomplete in that page. I don't fix as i am one of the missing ones :-) 86.254.162.206 (talk) 13:11, 28 May 2023 (UTC)