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

Kahn Jonas jonas.kahn at math.u-psud.fr
Fri Apr 1 12:00:49 PDT 2011


On Fri, 1 Apr 2011, valkyria at phmp.se wrote:

> Quoting Michael Williams <michaelwilliams75 at gmail.com>:
>
>> I wonder what would happen if you counted jigo as a loss for both 
>> players...
>
> Interesting, it would then avoid all lines that are likely to lead to a draw. 
> The problem is that it then not be able to see any difference between a jigo 
> and a loss which may be a problem in a position which is a clear jigo. What 
> happens deeper in the tree does not matter, if there is only a few jigo 
> candidate moves and many losing moves, the program will often pick a losing 
> move.

I had almost posted that on yesterday, but I did not know whether you
could disconnect the result for W and B. Otherwise, the easiest hack to
discourage jigo without making it like defeat is:
count jigo as victory for each player say 30% of the time, and as defeat
for W otherwise.

Jonas



More information about the Computer-go mailing list