Re: [computer-go] Pattern matching - example play

["Frank de Groot" <frank@xxxxxxxxxxxxx>]

From: "Heikki Levanto" <heikki@xxxxxxxxxxxxxxxxx>
Subject: Re: [computer-go] Pattern matching - example play


> Nice theory, but I find it pretty hard to believe, for many reasons.
> In a way that is a "typical chess programmers" approach to Go,

I try to be as far away from the chess approach as possible, because Go has
only one type of "piece" and they don't move and the search space is
massively larger.


> find rather simpleminded and quite unlikely to produce good results. Not
> because the searhc space is too large, but because I do not believe it
> is possible to produce a whole-board evaluation function that gives
> meaningfull results fast enough for a deep search.

Not when you use the right type of search.
Humans can do it, so why not a computer?

What might help is huge pattern libraries of all kinds to direct the search
and evaluate positions in terms of connectivity, shape, eye space etc.

_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/