> You'll probably enjoy an article I wrote last year on this theme: > http://dcook.org/compgo/article_the_problem_with_random_playouts.html
Hi Darren, yes, I particularly liked that you explained what you were trying to show and how you think your data supports this - one might agree or disagree, but at least it is fairly clear what one agrees or disagrees with. The concrete references also led me to earlier discussions that showed that the biases and blind-spots I was trying to highlight were already well known to this group. Unsurprisingly, really, but I hadn't seen these points emphasized much elsewhere (be it because the standard techniques mostly work, or because it isn't clear yet how they work and hence how to improve them). What I'm looking for is a kind of recent survey, summarizing all those well-known ideas and problems that aren't much talked about here anymore (unless a newcomer like myself re-encounters them for the first time;-). The field is probably still too much in flux to hope for a complete survey in the style of these two from 2001/2002 http://www.citeulike.org/group/5884/tag/survey But if someone knowledgable were to draft such a thing specializing on Monte-Carlo and other developments since 2002, as a technical report with yearly updates, that would be a great resource, would avoid repeat questions in this group, and could ultimately evolve into the next great survey paper.. Thanks, Claus _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
