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

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

From: "David Weiss" <dweiss51@xxxxxxxxxxxxxxxxx>
Subject: Re: [computer-go] Pattern matching - example play


> This has been pointed out before on this mailing list.
> The moves that are actually played are the results
> of many moves being rejected that are not seen in
> professional or strong amateur games.

Exactly.

This is what my pattern system does best.
I will not choose the moves that are not seen in pro games because it has
not made a list of those moves as being "popular in pro games" (in the
context of the other patterns on the board that ARE seen in pro games)


> A program that attempts to predict pro moves will
> have a hard time when the opponent makes moves that
> would be terrible against a strong player.

I think a program that predicts pro moves is not a Go program.
There is a lot of confusion based on a lot of FUD about what my pattern
system does.
It's a out-of-proportion Joseki-dictionary of which I test the performance
by looking how many pro moves are predicted, which is a direct measure of
how closely the pro joseki is "followed".



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