2009/7/22 Andrés Domínguez <[email protected]>

> 2009/7/20 Stefan Kaitschick <[email protected]>:
> > Ofcourse they can know. They just have to check for it.
> > Those programs that do well against mirror go probably all do check for
> it.
>
> I think a strong MCTS could find the lines that make mirror Go
> useless. Maybe MF plays lines that brake mirror or related capture
> races because this lines have more probability of winning.


I don't know if there is any evidence that MFGO plays this any better than
ZEN,  unless it's stronger than ZEN in general.   If MFGO is stronger than
Zen, then it should be no surprise.   Otherwise ...

In order for this to be scientific, you need a reasonable number of game
examples.    I can imagine that a program could look very foolish in one
game, then brilliant in another and this might be totally dependent on
whether it blundered into the right kind of position where the normal
algorithm would punish the mirror play.

So what is the basis for saying that MFGO easily handles this  and Zen is
totally clueless about mirror go?   It could be true, but is there anything
more than anecdotal evidence?


- Don




>
>
> Andrés
> _______________________________________________
> computer-go mailing list
> [email protected]
> http://www.computer-go.org/mailman/listinfo/computer-go/
>
_______________________________________________
computer-go mailing list
[email protected]
http://www.computer-go.org/mailman/listinfo/computer-go/

Reply via email to