[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gnugo-devel] move valuation

From: Gunnar Farneback
Subject: Re: [gnugo-devel] move valuation
Date: Thu, 15 Nov 2001 17:30:36 +0100
User-agent: EMH/1.14.1 SEMI/1.14.3 (Ushinoya) FLIM/1.14.2 (Yagi-Nishiguchi) APEL/10.3 Emacs/20.7 (sparc-sun-solaris2.7) (with unibyte mode)

Arend wrote:
> Yes, I read this. I think technically more correct should be the following
> formula:
> [...]
> (If you think this is worth trying out I can send it as a patch of course.)

I think it's worth trying. At the very least we can see what breaks
and try to determine why.

> > > (Like playing into a knight's move:
> > > .X..
> > > ..*O
> > > ..X.
> > > ....
> > > ----
> > > which is bad aji keshi.)
> > [...]
> But I can hardly imagine a territorial_value estimation that will give
> a negative result here before you add the X answer (and a negative value
> would be correct here).

That's not entirely unrealistic, but as long as it's small enough that
some better move locally is valued more, it suffices. By the way, that
move seems to have fallen out of favor recently. My working copy
values it to 2.95 points in test case arend:16.

> Another example I have in mind is the wedge into an iken tobi: GnuGo plays
> it too often (and is afraid of it as well), as it disconnects the two
> stones, thus creating two weak groups. But of course you cannot estimate
> the result of the wedge without considering the (usually bad) result:

Yes, this one is relly tricky. It may be possible to workaround the
problem in the influence function, but it's probably quite hard.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]