The MoGo programmer who answered questions after the match (Olivier
Teytaud) did state that MoGo no longer used UCT. He gave a one-line
statement of the reason they switched, which I did not follow.
Bob
On Aug 12, 2008, at 7:00 PM, Jim O'Flaherty wrote:
All,
Can anyone detail the design of the version of Mogo that beat the
professional? Or is there a web-page where at least the general
approach has been described? Is the information even public? I am
not seeing the the implementation details, just the overall design
and general strategies. However, if the implementation details are
available, I would love to see those as well.
I am confused around what Mogo used. Was it Monte Carlo only, UCT
only, Monte Carlo integrated with UCT, RAVE, etc? I have read
through all of the recent emails, and I have not been able to get a
clear picture of it's design. Mogo at one timed used Monte Carlo and
UCT. I read an email that the one that played the pro and won did
not use UCT at all. However, I thought the massive tests that Don
did awhile back showed that MC did not scale very well, but MC + UCT
did.
And what language/platform is Mogo written in; C/C++, Java,
Assembly, PHP, etc.? And how did the language/platform choice impact
the overall efforts; speed them up, slow them down, complicate/ease
creating the parallelism on the super computer, etc.?
So, I am now confused precisely what method or methods were used and/
or integrated to produce the current scalable version of Mogo. I
want to know these details so I can at least get a better sense of
what actually occurred with the win. I don't care near as much about
the hardware as I do the software architecture and design.
Thank you,
Jim
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
