[computer-go] Experiments with UCT

[Chris Fant]

At first I didn't see the point, but maybe if you get several players,
you can once and for all put to rest the question of what the optimal
7x7 line is.  I think you will see many repeat games along the way.


On 7/27/06, Don Dailey <drd at mit.edu> wrote:
> If someone is willing to run a copy of the server at 7x7,  we can try it
> for a few days - but I don't want to distract too much from the 9x9
> server.   Do you think there would be serious interest to run it
> somewhere for a few days?
>
> - Don
>
>
>
> On Thu, 2006-07-27 at 21:02 -0400, Jason House wrote:
> > 7x7 would be very interesting.  Especially considering that 7x7 go is
> > semi-solved.
> >
> > Jason.
> >
> > Don Dailey wrote:
> > > On Tue, 2006-07-25 at 17:46 -0400, Chris Fant wrote:
> > >> Are you experimenting with these parameters for 5x5, 6x6, 7x7 to get
> > >> an idea of how they scale with boardsize so that you can make a good
> > >> guess at what they should be on a board that's harder to optimize,
> > >> like 9x9?
> > >>
> > >
> > > No.  I'm only trying to optimize for 9x9 right now.   My program can
> > > play on any square board but I have no idea what would happen on bigger
> > > or smaller boards.
> > >
> > > Maybe we set up a 7x7 CGOS server some time and see what happens?
> > >
> > > - Don
> >
>
> _______________________________________________
> computer-go mailing list
> computer-go at computer-go.org
> http://www.computer-go.org/mailman/listinfo/computer-go/
>