Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!news.alpha.net!uwm.edu!cs.utexas.edu!news.sprintlink.net!EU.net!ieunet!curia!odyssey.ucc.ie!colin
From: colin@odyssey.ucc.ie (Colin McCormack)
Subject: Re: Alopex
Message-ID: <D676ux.D2C@curia.ucc.ie>
Sender: usenet@curia.ucc.ie
Organization: University College Cork, Ireland
X-Newsreader: TIN [version 1.2 PL2]
References: <D65K7n.1GC@pts.mot.com>
Date: Wed, 29 Mar 1995 10:37:44 GMT
Lines: 28

Dale Mason (dmason@pts.mot.com) wrote:
: I did not read the FAQ, because I thought this would not be there and have tried
: looking at some books to no avail, but there is a learning Algorithm that is of
: recent exposure called "ALOPEX".  Anyone heard of it?  I am not concerned with its
: advantages and disadvantages, but am in desparate need of finding the Algorithm on 
: how it works. 

look at:
Unnikrishnan, K.P.      Learning in Connectionist Networks using the Aloplex
Algorithm. in neuroprose (see FAQ) as unni.aloplex.ps.Z
and
Venugopal, K.P. Learning in Connectionist Networks using the Aloplex Algorithm
also in neuroprose (subdir Thesis) as venugopal.thesis[1-7].ps.Z

the author also distributed source code a while back, you might drop him a 
line, last known address was: venu@pixel.mipg.upenn.edu

btw has anyone looked at the advantages and disadvantages of this algorithm ?
imo correct learning rule parameter selection seems to be a little sticky.
				regards,
					colin.
					
?#"#%^#$&$%&&$%&#"$&"%&"&"^%""^^&*)%&*!%$#"(()*&%**$"#_)+~)*(&"!%!!"^%&%&&&%#&%
Colin McCormack,                         
Dept. of Computer Science,            WWW: http://odyssey.ucc.ie/www/colin.html
University College Cork,	      E-mail: colin@odyssey.ucc.ie
Ireland.                              Phone: 353-21-902795
!/';./\3"^^#$&%*#$^(*($"#!$%/!$&^&{}"{["&^"^"^)&*(!+_+!_%~!%)!$#(!(%^)%(^%()%(
