Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!Germany.EU.net!zib-berlin.de!gs.dfn.de!fauern!rrze.uni-erlangen.de!rachael.franken.de!hub-n.franken.de!ark.franken.de!ralf
From: ralf@ark.franken.de (Ralf W. Stephan)
Subject: Re: Back Prop, according to Caudill/Butler
Message-ID: <1995Feb3.105938.10243@ark.franken.de>
Organization: his desk writing an article
Date: Fri, 3 Feb 1995 10:59:38 GMT
References: <3g498u$60k@dcsun4.us.oracle.com> <3g4gn9$t69@cantaloupe.srv.cs.cmu.edu> <3g6abs$9ff@agate.berkeley.edu> <3gja6h$o7l@dcsun4.us.oracle.com>
X-Newsreader: TIN [version 1.2 PL2]
Lines: 12

gbarish@netcom.com writes:
> Any idea what kind of hardware the groups you mention run their
> high-param simulations under?  Are they using neurocomputers or
> add-in boards?  And on what platforms?

In his papers, Dr. Tony Robinson for example mentions a transputer
machine with 64(?) nodes for training a recurrent net.

The papers are available on svr-ftp.eng.cam.ac.uk//pub/reports


ralf
