The people with stronger programs and more full-board experience will be better positioned to comment on this. I'll say that the two styles both need a lot of tweaking, making it hard to establish a fair test between them.
It's good to be able to match a 3x3 pattern very quickly. Since there aren't so many of them, it makes sense to build a lookup table. - Dave Hillis -----Original Message----- From: Jason House <[EMAIL PROTECTED]> To: computer-go <computer-go@computer-go.org> Sent: Mon, 21 Jul 2008 10:27 pm Subject: Re: [computer-go] Incremental move weights On Jul 21, 2008, at 10:03 PM, [EMAIL PROTECTED] wrote: I've tried both MoGo and CrazyStone style playouts. I find MoGo style playouts easier to work with. YMMV. Besides ease of implementation, what other trades did you notice between the two methods? I'd expect Crazy Stone's method to be slower. Is it stronger? My MoGo style heavy playouts are about 4 times slower than my light playouts. That's not too bad, considering... Were there any tricks to minimize the slowdown? -----Original Message----- From: Jason House <[EMAIL PROTECTED]> To: computer-go <computer-go@computer-go.org> Sent: Mon, 21 Jul 2008 8:45 pm Subject: Re: [computer-go] Incremental move weights On Jul 21, 2008, at 7:54 PM, [EMAIL PROTECTED] wrote: I use proximity in the heavy playouts themselves. What kind of proximity heuristic do you use? Remí' s paper implies a different weight for many many points on the board. I think MoGo uses an alternate approach of examining the local neighborhood and then considering a tenuki. How much of a slow down did you observe when you went to heavy playouts? I think most (all?) people do this. I have a precalculated table with the 3x3 and 5x5 neighbors for every position on the board. - Dave Hillis - ----Original Message----- From: Jason House <[EMAIL PROTECTED]> To: computer-go <computer-go@computer-go.org> Sent: Mon, 21 Jul 2008 7:32 pm Subject: [computer-go] Incremental move weights I'm starting heavy plyouts, with variable move selection weights. The proximity heuristic seems like a performance killer since most weights would require an update with each move. How do others handle this? Is proximity reserved for the search tree? How do others store data for rapid lookup? Sent from my iPhone _______________________________________________ computer-go mailing list [EMAIL PROTECTED] http://www.computer-go.org/mailman/listinfo/computer-go/ The Famous, the Infamous, the Lame - in your browser. Get the TMZ Toolbar Now! _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ = _______________________________________________ omputer-go mailing list [EMAIL PROTECTED] ttp://www.computer-go.org/mailman/listinfo/computer-go/ The Famous, the Infamous, the Lame - in your browser. Get the TMZ Toolbar Now! _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ = _______________________________________________ omputer-go mailing list [EMAIL PROTECTED] ttp://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/
