Newsgroups: comp.ai.neural-nets
From: jimmy@ecowar.demon.co.uk (Jimmy Shadbolt)
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.kei.com!bloom-beacon.mit.edu!gatech!swrinde!pipex!demon!ecowar.demon.co.uk!jimmy
Subject: Re: MLP vs Projection Pursuit 
Distribution: world
References: <1994Dec6.174703.10427@newsgate.sps.mot.com>
Organization: Econostat
Reply-To: jimmy@ecowar.demon.co.uk
X-Newsreader: Simple NEWS 1.90 (ka9q DIS 1.21)
Lines: 24
Date: Thu, 8 Dec 1994 10:32:59 +0000
Message-ID: <786882779snz@ecowar.demon.co.uk>
Sender: usenet@demon.co.uk

In article <1994Dec6.174703.10427@newsgate.sps.mot.com> chance@ae.sps.mot.com writes:

>Warren Sarle (saswss@hotellng.unx.sas.com) wrote:
>
>: In article <66979.alison@atp.biochem.usyd.edu.au>, 
> <alison@atp.biochem.usyd.edu.au> writes:
>: |> Given a nonparametric regression problem in which there are multiple,
>: |> continuous-valued inputs and outputs (i.e., multivariate / multiresponse)
>: |> what are the pros and cons of using a multilayer perceptron (with sigmoid
>: |> hidden units and linear output units) and projection-pursuit regression
>: |> (with multuple outputs)....
>
.. but I've never heard heard of it and am just 
wondering whats the general idea...
>

        Girosi and Poggio et al have shown that their RBFN can be
        seen as projection pursuit-type regression methods. Check
        out ftp publications.ai.mit.edu.

        Good luck

        Drago

