Re: [computer-go] Re: Amsterdam 2007 paper

2007-06-11 Thread Sylvain Gelly
Hi John, You mean they are positions in which humans passed and scored the game? And you're just testing if the simulation correctly converts all territories into 1-point eyes, and kills all invasions? Actually MoGo (10k simulations) labelled the positions, then a human checked the labels. The

Re: [computer-go] Re: Amsterdam 2007 paper

2007-06-11 Thread John Tromp
hi Sylvain, > Figure 3 in your UCT paper shows the accuracy of different simulation policies. > Could you repeat these experiments for accuracy of win/loss determination only? Actually the labelled positions are rather end game positions, and are labelled as 0/1 (loss/win). So we already are i

[computer-go] CGF special meeting 2007 result

2007-06-11 Thread Hiroshi Yamashita
CGF(Computer Go Forum) special meeting(Computer Go tournament) was held in Japan, 9th,10th June. There were 7 programs include GnuGo as a guest. And GnuGo and Aya won the tournament by 5-1 score. "agouti" was a new program written by Kenichiro Yasuhara. GnuGo won this tournament for 4 years in a

Re: [computer-go] Re: Amsterdam 2007 paper

2007-06-11 Thread Sylvain Gelly
Hello John, Thank you for your interest. Figure 3 in your UCT paper shows the accuracy of different simulation policies. Could you repeat these experiments for accuracy of win/loss determination only? Actually the labelled positions are rather end game positions, and are labelled as 0/1 (loss/

Re: [computer-go] Re: Amsterdam 2007 paper

2007-06-11 Thread John Tromp
hi Sylvain & David, Figure 3 in your UCT paper shows the accuracy of different simulation policies. Could you repeat these experiments for accuracy of win/loss determination only? So for each test position, you determine if it's won or lost under perfect play, and then see how close each policy g

Re: [computer-go] Congratulations to GNU and to MoGoBot19!

2007-06-11 Thread Chris Fant
Could a reasonably skilled human play a 19x19 games with 1 second per move? Let's say he starts with 5 seconds and 1 second is added to his clock after each move. I'm not asking for a high quality game, but could he manage a non-silly plausible game using cgoban on KGS? It seems like the stra

Re: [computer-go] OSS or Free Engines

2007-06-11 Thread Eduardo Sabbatella
I should add my open source project in the wiki page. (but first of all, I should finish uploading it and release the first version. :-) ) Take a look at: http://sourceforge.net/projects/javago Cheers Eduardo --- Jason House <[EMAIL PROTECTED]> escribió: > Joshua Shriver wrote: > > Anyone have