Good catch Yamato. I think the idea is that they're trying to calculate the true variances rather than the sample variances. It's true that q_ur would probably give a better estimate than q_u or q_r alone. Of course, q_ur depends on beta, and as they calculate it, beta depends on q_ur.
It may be ok to use q_r since it should have the most samples, or just do as you suggest. Still, it'd be nice to see the authors of this respond with what they really do since what's stated there can't be right. On Feb 15, 2008 11:38 PM, Yamato <[EMAIL PROTECTED]> wrote: > I am very confused about the new UCT-RAVE formula. > The equation 9 seems to mean: > > variance_u = value_ur * (1 - value_ur) / n. > > Is it wrong? If correct, why is it the variance? > I think that the variance of the UCT should be: > > variance_u = value_u * (1 - value_u). > > Why cannot we use that? > > Anyway, can anyone write the pseudo-code of this algorithm? > > -- > Yamato > _______________________________________________ > computer-go mailing list > [email protected] > http://www.computer-go.org/mailman/listinfo/computer-go/ >
_______________________________________________ computer-go mailing list [email protected] http://www.computer-go.org/mailman/listinfo/computer-go/
