Re: [gnugo-devel] Patch: improve separation of similarly-valued moves

[Paul Pogonyshev]

Gunnar wrote:
> Paul wrote:
> > It must be quite simple, as long as we can tweak a new mode into
> > engine in which it would _not_ apply random shifts to move values,
> > but instead would report the scales of those shifts to the caller.
> > I haven't looked at the exact math involved, but I'm sure it can
> > be done strictly with theory of probability, not involving any
> > statistical methods of probability assessment.
> 
> It certainly can be computed exactly but the question is whether it's
> practical to do that. Monte-Carlo evaluation of the involved integrals
> could very well be the most reasonable solution.

Don't forget that all statistical methods, including Monte-Carlo, give
approximate results.  In other words, if we implement Monte-Carlo assessment,
we will have some random probability deviations, which are generally a bad
thing.

Paul