Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!ut-sally!im4u!rutgers!ucla-cs!zen!ucbvax!boulder.UUCP!mikek
From: mikek@boulder.UUCP (Mike Kranzdorf)
Newsgroups: comp.ai.digest
Subject: Re: Neural Networks & Unaligned fields
Message-ID: <2118@sigi.Colorado.EDU>
Date: Fri, 4-Sep-87 12:13:31 EDT
Article-I.D.: sigi.2118
Posted: Fri Sep  4 12:13:31 1987
Date-Received: Wed, 9-Sep-87 04:45:46 EDT
References: <277@ndmath.UUCP> <12331701930.42.LAWS@KL.SRI.Com>
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: boulder!mikek@boulder.colorado.edu (Mike Kranzdorf)
Organization: University of Colorado, Boulder
Lines: 9
Approved: ailist@stripe.sri.com


	The second reference above is correct, but fails to mention work
by Fukishima and Mozer.  These multi-layer networks are able to form
an internal distributed representation of a pattern on an input retina.
They demonstrate very good shift and scale invariance.  The new and
improved neocognitron (Fukishima) can even recognize multiple patterns
on the retina.

--mike					mikek@boulder.colorado.edu