On Tue, Oct 5, 2010 at 16:59, Erik van der Werf <[email protected]> wrote: > Lukasz brings up an interesting point. Winrate may not be the ideal > statistic for all situations. Maybe the average score (as used in most > early work on MC), soft-max, or a median tracker would be better for > some situations. > > Maybe a nice question for the academics: > > If you were free to keep track of a histogram for all possible scores > in each node (so you have everything from winrate at every possible > komi to simply the average score), then what would be the optimal > selection strategy? > > And if the above can be answered, what would be the minimum set of > statistics needed to maintain similar performance?
As Kahn said it is difficult to find anything theoretically since playouts are so different from the actual play. The problem is that the variance of the playout's results go down to zero much later (very very late endgame, or never) than the variance of score in actual game (whenever game is close to decided). So we can't use variance of playouts (what would be natural) to guide the game. > > Erik > > > On Tue, Oct 5, 2010 at 4:08 PM, Aja <[email protected]> wrote: >> I think what Łukasz meant of "score" is the exact game scoring (such as >> Black wins 10 points, then return 10, rather than 1). He might have greater >> ambition to change the current architecture of MCTS that only 0/1 is used >> for the outcome of the simulations. >> >> Aja >> >> ----- Original Message ----- From: <[email protected]> >> To: <[email protected]> >> Sent: Tuesday, October 05, 2010 9:34 PM >> Subject: Re: [Computer-go] Results from 19x19 Valkyria/1k H9 vs Valkyria/10k >> >> >>> Quoting Łukasz Lew <[email protected]>: >>> >>>> Your dynamic komi results are very convincing.. >>>> But shouldn't we just concentrate on maximizing score instead of >>>> winning rate in the beginning of the game? >>> >>> Maximizing winning rate means that the probability of having a score > 0 >>> at the end of the game is maximum. >>> >>> Exactly what do you mean with "maximizing score in the beginning of the >>> game"? >>> It is hard to estimate the score. Also in a game of go territory is not >>> everything. Aji and influence is also important. Win rate is as far as i >>> know the best way of capturing all these things in one measure that guides >>> search. >>> >>> Magnus >>> >>> >>> _______________________________________________ >>> Computer-go mailing list >>> [email protected] >>> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go >> >> _______________________________________________ >> Computer-go mailing list >> [email protected] >> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go > _______________________________________________ > Computer-go mailing list > [email protected] > http://dvandva.org/cgi-bin/mailman/listinfo/computer-go -- Łukasz _______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
