Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!lll-crg!lll-lcc!pyramid!oliveb!sun!lyang
From: lyang@sun.uucp (Larry Yang)
Newsgroups: comp.arch
Subject: Neurotic Nomenclature
Message-ID: <9138@sun.uucp>
Date: Mon, 10-Nov-86 19:43:47 EST
Article-I.D.: sun.9138
Posted: Mon Nov 10 19:43:47 1986
Date-Received: Tue, 11-Nov-86 06:29:39 EST
Distribution: net
Organization: Sun Microsystems, Inc.
Lines: 31

The concept of neural networks is fascinating when I
started reading about it.  Coming from combinational
biological sciences/electrical engineering, I find this
to be a unique opportunity to simulate the behavior of
natural organisms.  However, the more I study about
neural networks, the less it sounds like what it means.

Does the term "neural network" simply imply a network of
simple analog computing nodes, connected in a simple
"neural" fashion?  Or does it really emulate a neuron's
function, like making new connections and unmaking old
connections (i.e., learning and forgetting) and transmitting information 
digitally (actually, frequency encoded) and operating on it "analogly".
Is "neural network" just a fancy, grant-catching term, such as
"artificial intelligence" (oooo, I'm gonna get flamed for that... :-)





-- 
==============================================================================
                                         /-\   /\    |
-- Larry Yang                          __|_/    |    |
   Sun Microsystems, Inc.                |      |    | /--\ /\/-\/--\
   Mountain View, CA  94043            /-|\     \    / |  | ||  ||  |
   lyang@sun.COM                       \_/ \     \__/| /--/\/|  \/--| /
   {ihnp4|hplabs|seismo}!sun!lyang                 __|/           __|/
                                                  /  /           /  /
                            (a REAL signature)    \_/            \_/
==============================================================================