Re: [computer-go] Statistical significance (was: SlugGo vs ManyFaces, newest data)

[Matthew Saffell <saffell@xxxxxxxxxxx>]

Couldn't you just have them play two games with same opening, switching
roles for the set?  Then the grossly unfair positions would tend to wash
out more quickly.  I guess you would expect them to wash out anyway given
a long enough sequence of games since the unfair instances should crop up
randomly for either opponent in your original setup, but maybe repeating
the setup with roles reversed might allow for shorter tournaments for the
same level of confidence.

Matt


On Fri, 10 Sep 2004, Nicol N. Schraudolph wrote:

> Don,
> 
> > The  problem with this  is that  it tends  to equalize  opponents.  If
> > version  B is  stronger than  version A,  giving them  random starting
> > positions  will tend  to  give a  little  advantage to  the version  A
> > because now  the version  A will tend  to get some  starting positions
> > that are  heavily in his favor.
> 
> how about setting up symmetric random opening positions?  I.e., let Black
> play random moves, and White mirror them.
> 
> - nic
> 
> -- 
>     Dr. Nicol N. Schraudolph                 http://n.schraudolph.org/
>     Sonnenkopfweg 17
>     D-87527 Sonthofen, Germany
> 
> 
> 
> _______________________________________________
> 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/