[Computer-go] 7.0 Komi and weird deep search result

Vlad Dumitrescu vladdu55 at gmail.com
Fri Apr 1 01:35:57 PDT 2011


On Fri, Apr 1, 2011 at 10:28, <valkyria at phmp.se> wrote:

> The question is: should jigo be counted balanced for both players or should
> one perhaps bias the randomized winner for jigo playout. For example if the
> program is winning 60% at root then jigo would count as win in only 40% of
> the cases for the color of program in the playouts. This would perhaps drive
> the program more towards winning lines when it wins and towards jigo when it
> loses.

Just a thought: introducing jigo as a possible result creates a different
situation at a meta-strategic level. In a tournament situation, maybe I
absolutely need a win, then a jigo is the same as a loss. Or it doesn't
matter and then jigo is the same as a win.

best regards,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://computer-go.org/pipermail/computer-go/attachments/20110401/7dad2974/attachment.html>

More information about the Computer-go mailing list