IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

machine brains

ray scanlon rscanlon at wsg.net
Tue Sep 30 19:27:15 EST 1997



Gerben_Wierda at RnA.nl wrote in article
<01bccd31$d50075c0$eb7a61ce at asdf>...
> Alexander Solntsev <solntsev at wt.net> wrote:
> >ray scanlon wrote:
> >> 
> >> ...
> >> So I will go ahead and attempt to attempt to understand the brain
> >> as a neural net with the neurons as electrical constructs.
> >> 
> >I just want to point out that neural net is not the only highly
> >interconnected inherently parrallel computing device (entity).
> 
> I'd like to add that neural net implementations on digital computers
> are just 
> "data driven rule based systems in disguise". They circumvent the 
> 'programming problem' (just like genetic algorithms) and are
therefore
> very 
> efficient implementations, but the bottom line is that they do not
> differ 
> from any other rule based system in theoretical capability.

A "neural net implementation on a digital computer" should be viewed as
a simulation of a neural net constructed of neuromimes. It is not the
"thing" but a simulation of the "thing".

The human brain is also a neural net and may be simulated in exactly
the same way.

The human brain is a rule based system, the rules are established by
the DNA. The DNA  sets up the rules of synaptic modification and axonal
growth. It also establishes the rules that set up the original hard
wiring.

What's wrong with the theoretical capabilities of a rule based system?

ray

-- 

email: rscanlon at wsg.net

If you are interested in how the brain works, visit
http://www.wsg.net/~rscanlon/brain.html



More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net