It's not the same program as the one from 1995--maybe I should change the
name to avoid confusion. Argus is a program in development, and I'm
actually still not 100% sure it will be done in time for the tournament.
Sam
On Mon, Jun 23, 2008 at 1:04 PM, Nick Wedd <[EMAIL PROTECTED]> wrote:
> In m
UCB (and hence UCT) would treat the following sequences of wins (1)
and losses (0) the same:
01010101010101010101010101010101
Clearly, it would be better to favor the second sequence, because that
move has done more for us lat
Drake, P. & Chen, Y.-P. (2008) “Coevolving Partial Strategies for the
Game of Go”. In Proceedings of the 2008 International Conference on
Genetic and Evolutionary Methods, CSREA Press.
URL:
http://webdisk.lclark.edu/drake/publications/drake-gem2008-final.pdf
We've made some considerable imp
I think we all understand trying to get code done for an event. I gave
up any hope of that with SlugGo a little while ago. Now I am only
doing parameter tweaking (hunting in too many dimensions, all in the
dark).
Cheers,
David
On 25, Jun 2008, at 12:02 PM, Sam Gross wrote:
It's not the
Seems to be working now. I restarted the low level Many Faces, and added a
strong Many Faces.
Looks like the reference gnugo is still missing though.
David
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:computer-go-
> [EMAIL PROTECTED] On Behalf Of Seth Pellegrino
> Sent: Tuesda
My clients all got "irregular response from the server" at the first move of
the game, and it won't let them reconnect. Says they are already logged on.
SO maybe there is still some problem.
David
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:computer-go-
> [EMAIL PROTECTED] On
Thanks for that. It's nice to know that I'm not the only one.
David Doshay wrote:
I think we all understand trying to get code done for an event. I gave
up any hope of that with SlugGo a little while ago. Now I am only
doing parameter tweaking (hunting in too many dimensions, all in the
dark)
