[Top][All Lists]

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

Re: [gnugo-devel] tuning patch

From: Arend Bayer
Subject: Re: [gnugo-devel] tuning patch
Date: Sun, 13 Apr 2003 20:39:54 -0400 (EDT)

Paul wrote:

> i'm somewhat afraid to mark the move in AS7 as an antisuji. instead i
> added a large negative shape value (-10). however, M19 in the appended
> game is still valued 1.28 points. probably the best solution would be
> to revise antisuji moves as non-antisuji if they also have some vital
> move reason (e.g. OWL_ATTACK). the same may also hold for large negative
> shape values.

Maybe I am biased, but IMHO the best solution is to fix the territorial
move valuation of M19. This should be possible by pattern tuning in

<barriers.db pattern tuning advertisement>
The influence pattern database is pretty small, and the patterns tend to
be fairly general. This means that fixing or improving one of the patterns
there will help in many more cases than tuning patterns.db or endgame.db.
</barriers.db pattern tuning advertisement>

I am not against adding AS7, but it should be marked as a workaround for
a mistake in the influence module.


reply via email to

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