Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!travelers.mail.cornell.edu!news.kei.com!news.mathworks.com!tank.news.pipex.net!pipex!warwick!bsmail!usenet
From: Max Ratcliffe <m.j.ratcliffe@bris.ac.uk>
Subject: Re: Implementation of LM nnet training algorithm in MATLAB
Message-ID: <DF3xq7.H9M@uns.bris.ac.uk>
Sender: usenet@uns.bris.ac.uk (Usenet news owner)
Nntp-Posting-Host: pc19.aer.bris.ac.uk
Organization: University of Bristol, England
References: <436onm$rmq@taco.cc.ncsu.edu> <DF30Aw.Ep1@unx.sas.com> <43jus7$j9b@bubba.ucc.okstate.edu>
Date: Mon, 18 Sep 1995 15:30:07 GMT
Lines: 11

You say:
> 
>   Dr. Martin Hagan and his student Nidal are the first one who incorporated
>  Levenberg-Marquardt Algorithm into the Multilayer Network. For a reference,
> look at IEEE transaction on NN. IF you look into this paper, the LM algorithm
> is compared to Conjugate Gradient and Adaptive Learning. Accordingly, the LM
> has the fastest convergent speed for a small network. 

Which IEEE reference do you mean? Please give further details.

Thanks, Max
