Gian-Carlo Pascutto: <[EMAIL PROTECTED]>:
>Hideki Kato wrote:
>
>> What is "correct" move?  It has sense only for some artificial 
>> problems or very limited positions, and so, it cannot evaluate total 
>> performance of a program.
>> 
>
>This is true, but we are interested in search performance. So, it makes 
>sense to evaluate on those positions where the search makes the 
>difference, no?

I'm not sure but it's possible, as Don wrote, parallelizing mc/uct 
doesn't make its behavior change in the sense of search result will 
not change, ie, just decreases the efficiency of simulations.  
However, it's also likey that unbalanced long communication delay 
makes search results change.  Because the order of arrivals 
of results of simulations may change if unbalanced (long) delay 
exists.

Anyway, I'm implementing network parallel version of MC/UCT now and 
will make such things clearer.  

-Hideki
--
[EMAIL PROTECTED] (Kato)
_______________________________________________
computer-go mailing list
[email protected]
http://www.computer-go.org/mailman/listinfo/computer-go/

Reply via email to