Re: Why RBF classfier Vs sigmoidal MLP for Statistical classification?

Paul Victor Birke <[EMAIL PROTECTED]> wrote in message 
news:<[EMAIL PROTECTED]>...
> Greg Heath wrote:
> > [EMAIL PROTECTED] (UWS) wrote in message 
> > news:<[EMAIL PROTECTED]>...
> >   
> >>Could someone rationalize reasons to choose a RBF statistical
> >>classifier, instead of a sigmoidal MLP?
> >  
> > Sometimes RBFs work better...especially if the data is 
> > modeled better as a superposition of Gaussians.
> 
> even better Cauchy

Thick-tailed Cauchy has no variance. Theoretically,
Student-t may be more appropriate.

Hope this helps.

Greg