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

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

> Did you try to learn influence based on statistical information on how
> likely a point will become territory in a pro-game? That might actually be
> very interesting to see if that gives anything useful.


I have some scored games from Dave Dyer but I was planning to do it on a lot
of IGS scored games.

I gave it low priority because I think if anything is hard to do, then it's
predicting territory like that.

The reason I did not "learn" influence is that I consider it too simple to
need to do that.

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