Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!elroy.jpl.nasa.gov!aero!abbott
From: abbott@aerospace.aero.org (Russell J. Abbott)
Newsgroups: comp.ai.neural-nets
Subject: Re: Meta-Nets
Message-ID: <69267@aerospace.AERO.ORG>
Date: 21 Mar 90 17:15:11 GMT
References: <4670@cbnewsl.ATT.COM>
Reply-To: abbott@aero.UUCP (Russell J. Abbott)
Organization: The Aerospace Corporation, El Segundo, CA
Lines: 17

In article <4670@cbnewsl.ATT.COM> apr@cbnewsl.ATT.COM (anthony.p.russo) writes:
>We could, in principle and probably in actuality, train a net to
>teach other nets. I don't argue with that. However, the meta-net
>has been trained by some standard algorithm. What we have then,
>is a standard algorithm (say backprop) that is a teacher 
>of a teacher (the meta net) of a learner (other networks).

The original question was intended as a thought experiment whose purpose
was to examine the limits of neural nets.  If one could develop a
meta-net, then for a large class of problems the training phase would be
by-passed since the meta-net would be able to come up with the weights
directly.  It wouldn't be a trainer of the application net; it would
determine the weights for that net itself.  But is that reasonable: a
neural net system without the need for training?  If not, then why is a
meta-net impossible?
-- 
-- Russ abbott@itro3.aero.org