[computer-go] KO in Hashtable-UCT?

[Don Dailey]

On Sat, 2007-05-19 at 12:32 +0200, chrilly wrote:
> >
> > In the play-outs, I'm pretty sure infinite play-outs due to not using
> > superko are possible - even with the randomness.    But I have a limit
> > on the length of the play-out games because when you use heavy play-outs
> > the games can occasionally last for hundreds of moves.
> How do you evaluate a game which is stopped by reaching the length-limit?
> In UCT-Suzie I stop also when one side has a big material advantage 
> (captured much more stones). The length limit is related with this, because 
> when a big group is captured the empty points are afterwards filled up 
> again.

It's a rare occurrence that I have to stop a game early - but when I do
I simply score the board assuming all stones are alive.

Another thing you could do (if you're the anal type) is to finish the
game after a certain point with the superko rule turned back on.  You
would start having to track the keys again.   But this almost seems
silly because a random game is only a guess anyway.

> Chrilly
>