[Computer-go] Semeais

[terry mcintyre]

Imagine that the output of an analyzer were a small program snippet which dealt 
with all the various aspects of the semeai.


It would not only suggest the correct response for any move which altered the 
status of the semeai, but would deal correctly with move n+1, n+2, ... 

Call it an "autonomous agent."
 Terry McIntyre <terrymcintyre at yahoo.com>


Unix/Linux Systems Administration
Taking time to do it right saves having to do it twice.




________________________________
From: Stefan Kaitschick <Stefan.Kaitschick at Hamburg.de>
To: computer-go at dvandva.org
Sent: Fri, January 14, 2011 12:31:18 PM
Subject: Re: [Computer-go] Semeais

Move-response information fits perfectly into the mc flow. But it doesn't fully 
reflect the logic of a semeai with libs, shared libs, lib extensions, approach 
moves, and eyes of various sizes. So it can only be of limited value.
Incremental updating of information sounds like a idea good though. That might 
even eliminate or atleast reduce the need for top-level analysis.
In any case, just improving the playouts won't be good enough. Semeai 
information must be available in the tree to reduce useless permutations.

Stefan

> It is expensive to solve semeai during every playout, but what if semeai 
>solutions were done once per game move (or even less frequently), and the 
>information obtained were then amortized over many playouts?
> 
> The top-level analysis would solve the semeai, and create a self-updating set 
>of move-response triggers. The tricky bit is that the indicated behavior must 
>correctly adapt to all possible moves during a complete playout, in a manner 
>which does not interfere with other simultaneous playouts.
> Terry McIntyre <terrymcintyre at yahoo.com>

_______________________________________________
Computer-go mailing list
Computer-go at dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://computer-go.org/pipermail/computer-go/attachments/20110114/04fdbe06/attachment.html>