Re: [computer-go] EGC2007
I think CrazyPattern does not do any simulation at all. It just predict the move according pattern lib and Generalized Bradley-Terry model. Remi: it would be nice, if one could change the amount of simulations, altough I guess it has to be rather small so that those several hundred of games mentioned in your paper can be processed in reasonable time. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] EGC2007
I believe the original paper used 32 simulations from each point as part of its pattern... But the spirit of it (as I understand it) really is to bias the 1-ply move selection at the start of MC searches. (In Crazy Stone, just 3x3 patterns are used at deeper ply?) elife wrote: I think CrazyPattern does not do any simulation at all. It just predict the move according pattern lib and Generalized Bradley-Terry model. Remi: it would be nice, if one could change the amount of simulations, altough I guess it has to be rather small so that those several hundred of games mentioned in your paper can be processed in reasonable time. ___ 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/
Re: [computer-go] EGC2007
Jason House wrote: I believe the original paper used 32 simulations from each point as part of its pattern... But the spirit of it (as I understand it) really is to bias the 1-ply move selection at the start of MC searches. (In Crazy Stone, just 3x3 patterns are used at deeper ply?) elife wrote: I think CrazyPattern does not do any simulation at all. It just predict the move according pattern lib and Generalized Bradley-Terry model. Crazy Patterns does run 63 simulations. But there is no tree search, and I do not believe that more simulations would make it play any better. As Cai wrote, Crazy Patterns simply computes a probability distribution with the generalized Bradley-Terry model. It runs a few simulations anyways, because Monte-Carlo owner is one feature of the Bradley-Terry model. But there is no tree search at all. Rémi ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] EGC2007
Erik van der Werf wrote: It might be interesting to see how Rémi Coulom's move predictor does on these positions. Erik You can now donwload it as a GTP engine from the page of the paper: http://remi.coulom.free.fr/Amsterdam2007/ Rémi ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] EGC2007
On 8/10/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I organized side event Predict Professinal Moves at European Go Congress 2007. I made a brief and shallow summary (without any analysis) of the results and decided to post it here---in case someone is actually interested ;). The side event: http://users.tkk.fi/eseurane/EGC2007/experiment.pdf The summary: http://users.tkk.fi/eseurane/EGC2007/summary.pdf -Esa It might be interesting to see how Rémi Coulom's move predictor does on these positions. Erik ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] EGC2007
Erik van der Werf wrote: On 8/10/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I organized side event Predict Professinal Moves at European Go Congress 2007. I made a brief and shallow summary (without any analysis) of the results and decided to post it here---in case someone is actually interested ;). The side event: http://users.tkk.fi/eseurane/EGC2007/experiment.pdf The summary: http://users.tkk.fi/eseurane/EGC2007/summary.pdf -Esa It might be interesting to see how Rémi Coulom's move predictor does on these positions. Erik I'll make it available on my web page as a gtp engine, with analyze commands to show the probability distribution and move order. Rémi ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] EGC2007
uh, never mind. i should have looked a little more closely at the situation. :) s. - Original Message From: [EMAIL PROTECTED] [EMAIL PROTECTED] To: computer-go computer-go@computer-go.org Sent: Friday, August 10, 2007 8:34:26 AM Subject: [computer-go] EGC2007 I organized side event Predict Professinal Moves at European Go Congress 2007. I made a brief and shallow summary (without any analysis) of the results and decided to post it here---in case someone is actually interested ;). The side event: http://users.tkk.fi/eseurane/EGC2007/experiment.pdf The summary: http://users.tkk.fi/eseurane/EGC2007/summary.pdf -Esa ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ Yahoo! oneSearch: Finally, mobile search that gives answers, not web links. http://mobile.yahoo.com/mobileweb/onesearch?refer=1ONXIC ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
