Surely, the position between Zen and Pachi doesn't qualify :-)
As for a human-weaknesses database, my advice to any programing team
would be: just mind your own blind spots.
Stronger bots will need to reduce the current plague of refighting every
fight in every branch of the tree.
This gets even worse with several local fights at once. The useless
permutations between fights create a nightmare that cannot be overcome
with speed alone.
But the future solutions to this will surely create new problems
familiar to humans: false generalisations.
Stefan
To clarify:
I did not mean artificial positions
but positions reached from the empty board
in "normal" games, for instance between
human players or in those between humans
and bots or between bots and bots.
(Answer is also meant with respect to
the proposal by Oliver Lewis.)
One of the ideas with such a data base is to
use it when preparing a bot for a high-calibre
match against a top human. (I know it may be
still 25 years in future - but the time will
come.)
Ingo.
-------- Original-Nachricht --------
Datum: Tue, 10 May 2011 10:06:30 -0400
Von: steve uurtamo<[email protected]>
An: [email protected]
Betreff: Re: [Computer-go] Congratulations to Zen!
given the random games that were recently played, it's in fact most go
positions.
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
