[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gnugo-devel] a capturing race
From: |
Gunnar Farnebäck |
Subject: |
Re: [gnugo-devel] a capturing race |
Date: |
Fri, 09 Jul 2004 04:11:24 +0200 |
User-agent: |
EMH/1.14.1 SEMI/1.14.3 (Ushinoya) FLIM/1.14.2 (Yagi-Nishiguchi) APEL/10.3 Emacs/21.3 (sparc-sun-solaris2.9) MULE/5.0 (SAKAKI) |
Doug wrote:
> I think there are two mistakes here. The main one is misreading the
> capturing race at the bottom. But even aside from that, it doesn't seem
> wise for Black to go for a big exchange when a) Black is ahead and b) the
> profit side is uncertain. It seems like this mistake could be avoided even
> without fully understanding the capturing race.
There are two unsolved problems here:
1) Reliable estimation of uncertainty.
2) Integration of uncertainty estimates in the move valuation.
Advances in this area would certainly be welcome.
> And finally, on the subject of the capturing race, what does it mean for
> dragon_status to be critical if one of the moves is PASS? This from a few
> moves later:
Off the top of my head it means that the owl reading has failed to
analyze attack and defense consistently. That is of course a failure.
/Gunnar
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- Re: [gnugo-devel] a capturing race,
Gunnar Farnebäck <=