Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!yale!mintaka!bloom-beacon!eru!hagbard!sunic!dkuug!daimi!fodslett
From: fodslett@daimi.aau.dk (Martin Moller)
Newsgroups: comp.ai.neural-nets
Subject: Pre-print about scaled conjugate gradient available.
Message-ID: <1990Nov27.144401.10611@daimi.aau.dk>
Date: 27 Nov 90 14:44:01 GMT
Sender: fodslett@daimi.aau.dk (Martin Moller)
Organization: DAIMI: Computer Science Department, Aarhus University, Denmark
Lines: 41

Hello everybody,

The pre-print about scaled conjugate gradient, I wrote a note about
a while ago, is now available by ftp. A short abstract is:

	A supervised learning algorithm (Scaled Conjugate Gradient, SCG)
	with superlinear convergence rate is introduced. SCG is fully
	automated including no user dependent parameters and avoids a time 
	consuming line search, which other conjugate gradient algorithms use 
	in order to determine a good step size.
	
The following procedure should give you a copy of the paper:

	ftp daimi.aau.dk
	Name (daimi.aau.dk: (username)): anonymous
	Password: anything (your username)
	ftp>cd pub
	ftp> bin
	ftp>get scg.paper.Z
	ftp>quit


uncompress the paper when received by:

	uncompress scg.paper.Z

The paper can now be printed out on any postscript printer by:

	lpr -P<printername> scg.paper

The paper is 21 pages long.


Please send any questions or comments to:

	fodslett@daimi.aau.dk

	Martin F. Moller
	Computer Sci. Dept.
	University of Aarhus
	Denmark