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!europa.eng.gtefsd.com!howland.reston.ans.net!news.sprintlink.net!demon!ecowar.demon.co.uk!jimmy
Subject: Re: What NN is the best for control application? 
Distribution: world
References: <Cz2FMz.rE@osuunx.ucc.okstate.edu>
Organization: Econostat
Reply-To: jimmy@ecowar.demon.co.uk
X-Newsreader: Simple NEWS 1.90 (ka9q DIS 1.21)
Lines: 31
Date: Wed, 23 Nov 1994 10:09:49 +0000
Message-ID: <785585389snz@ecowar.demon.co.uk>
Sender: usenet@demon.co.uk

In article <Cz2FMz.rE@osuunx.ucc.okstate.edu> fmeng@master.ceat.okstate.edu writes:

>Hi,
>        The are a lot of NN architecture around now such
>as Radial Basis NN, Recurent NN, Feedforword NN, Modular NN 
>and ...
>        Which methods is the most accurate and most suitable
>for control application? 
     ^^^^^^^^^^^^^^^^^^^
     Can you be more specific?

>        To my knowledge, feedforward NN using 
>Marquardt-Levenberg method is the fastest in training. But,
>is Modular Neural Network faster than it?
    ^^^^^^^
    This is not FAQ type of network.
>What about others, eg Radial Basis and Recurent NN and etc?

        Radial basis are my favourite - you can *prove* some things
        if you use them (certain forms of stability, model order etc).
>
>Fun Meng Hock
> 
>

        Cheers

        Drago



