RE: [gnugo-devel] Re: [computer-go] SlugGo v.s. Many Faces

["David Fotland" <fotland@xxxxxxxxxxxxxxx>]

This is a great result.  At the Computer Games conference in Israel there
were several
programs that did playouts, and they were surprisingly strong.  

I'll add this to Many Faces and see it makes a big difference.

David

> -----Original Message-----
> From: computer-go-bounces@xxxxxxxxxxxxxxxxx 
> [mailto:computer-go-bounces@xxxxxxxxxxxxxxxxx] On Behalf Of Darren Cook
> Sent: Wednesday, August 25, 2004 3:31 PM
> To: computer-go
> Subject: Re: [gnugo-devel] Re: [computer-go] SlugGo v.s. Many Faces
> 
> 
> >>It is interesting that 16 moves of once-branched play is worth 
> >>something like 3 stones. We will continue to gather 
> statistics to find 
> >>out.
> > 
> > This form of search, deep analyses of single lines of play, I have 
> > called "playout analysis" 
> > <http://satirist.org/learn-game/inspire/playout.html>,
> > by analogy to the "rollout analysis" done by backgammon programs.
> 
> Nice to have a name for it. This is one of the ideas I've 
> used in my own 
> program after observing Many Faces: if you set it to play 
> itself out to the 
> end of the game on one of the low levels it takes just a 
> couple of seconds 
> but gives a more accurate score than pressing F9 to get 
> current score estimate.
> 
> I.e. "stupid, narrow but deep" beats "sophisticated but shallow".
> 
> The times it fails are when a non-obvious tesuji (e.g. a 
> throw-in or making 
> an empty triangle) is a considerably better move because it 
> lives/kills a group.
> 
> Darren
> _______________________________________________
> computer-go mailing list
> computer-go@xxxxxxxxxxxxxxxxx 
> http://www.computer-go.org/mailman/listinfo/computer-go/
> 

_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/