>> It is interesting that Crazystone, Zen (maybe) and Many Faces consider
>> patterns on the whole board during playouts, while Mogo just considers
>> them next to last move, and Fuego just next to the last 2 moves.
> 
> They consider patterns on the whole board, but also multiply values of
> spatial patterns with other weights, among others "contiguity" with very
> high value. So in practice, nearby pattern matches are very strongly
> preferred anyway.

Yes, that is fine. Where I thought the advantage might come is when
there are no pattern moves matching the next move, and it has to go
further down the list and choose a random move. A pattern move leftover
from another part of the board might be better than a random move, and,
on average, is unlikely to be worse.

>> I guess moving to whole board patterns doesn't actually add much CPU,
>> because they are only 3x3: a list of matching patterns can be
>> maintained, and after each move only a few points need to be considered
>> for new pattern matches.
> 
> Yes, but then you need to choose from a probability distribution over
> all the moves, which actually does slow things down *quite* a lot.
> But apparently the slowdown is worth it.

That is the Crazy Stone way? I imagined Fuego's way: just have a list of
moves and choose one of them randomly.

Darren



-- 
Darren Cook, Software Researcher/Developer

http://dcook.org/gobet/  (Shodan Go Bet - who will win?)
http://dcook.org/work/ (About me and my work)
http://dcook.org/blogs.html (My blogs and articles)
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go

Reply via email to