Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!solo!
From: ian@ponder.csci.unt.edu (Ian Parberry)
Newsgroups: comp.ai.neural-nets
Subject: Re: Postulates on the number of neurons
Message-ID: <1990Dec11.233435.863@solo.csci.unt.edu>
Date: 11 Dec 90 23:34:35 GMT
References: <1990Dec11.174921.26838@cs.columbia.edu>
Sender: ian@dept.csci.unt.edu (Ian Parberry)
Followup-To: comp.ai.neural-nets
Organization: University of North Texas, Denton
Lines: 25

In article <1990Dec11.174921.26838@cs.columbia.edu>
fahn@hudson.columbia.edu (Paul Fahn) writes:
>It seems to me that (computer-based) neural net researchers too often
>presume that biological organization is somehow optimal. It may be that
>the brain represents an enormous waste of resources, and that 
>"intelligent behavior" can be achieved with far fewer neurons, if only
>they were organized more "intelligently". (or, equivalently, that higher
>intelligence could be achieved with the same number of neurons.)

I used to take it for granted that we could hope to beat biology
(and I was dumb enough to say it in print, see the Intro to my
``Primer on the Complexity Theory of Neural Networks'', 
in ``Formal Techniques in Artificial Intelligence: A Sourcebook'',
R. B. Banerji (Ed.), in series Studies in Computer Science and
Artificial Intelligence, Vol. 6, pp. 217-268, Elsevier, 1990).
However, I met a biologist at the NIPS workshop who argued the opposite:
that perhaps brains are the best way of doing things within reasonable
constraints (speed, amount of hardware, volume, power, robustness,
energy supply, fault-tolerance, etc.)

Now I'm not so sure.
_____
Ian Parberry  ian@dept.csci.unt.edu  Dept. of Computer Science,
Univ. of North Texas, P.O. Box 13886, Denton, TX 76203-3886
A man with 1 watch knows what time it is.  A man with 2 watches is never sure.