ProDeo
Would you like to react to this message? Create an account in a few clicks or log in to continue.
ProDeo

Computer Chess
 
HomeHome  CalendarCalendar  Latest imagesLatest images  FAQFAQ  SearchSearch  MemberlistMemberlist  UsergroupsUsergroups  RegisterRegister  Log in  

 

 Evidence That NNs Work Best With Multiple Modules

Go down 
AuthorMessage
TheSelfImprover

TheSelfImprover


Posts : 3017
Join date : 2020-11-18

Evidence That NNs Work Best With Multiple Modules Empty
PostSubject: Evidence That NNs Work Best With Multiple Modules   Evidence That NNs Work Best With Multiple Modules EmptySun Dec 12, 2021 1:22 pm

Researchers at Fujitsu an MIT have shown that for understanding images, NNs work best when broken into multiple modules, each of which does a separate part of the work.

https://www.techradar.com/news/boffins-unveil-artificial-intelligence-that-thinks-just-like-we-do

It seems obvious to me that the same will apply to chess: one NN to do everything is clearly not the way. This partially explains why NNs trained on several order of magnitude more games than any human will ever see are unable to beat top humans at ply 1.

Tibono likes this post

Back to top Go down
 
Evidence That NNs Work Best With Multiple Modules
Back to top 
Page 1 of 1
 Similar topics
-
» Joan Baez a multiple person?!
» I wonder what is the reason that so many cores work for stockfish

Permissions in this forum:You cannot reply to topics in this forum
ProDeo :: Computer Chess-
Jump to: