X-Message-Number: 4289
Date: Tue, 25 Apr 1995 09:54:50 -0700
From: John K Clark <>
Subject: SCI.CRYONICS Neural Nets and Computers

-----BEGIN PGP SIGNED MESSAGE-----

In #4287  (Thomas Donaldson) Wrote: 

		>It's not obvious to me that neural nets are computers at all.

At first I thought I had no real disagreement with Professor Donaldson, 
just a different use of words. By "computer" I mean a determinist 
Turing Machine and I figured he was just using a narrower definition. 
I have no problem with that, but then he wrote:
		  
		>A simulation of a neural net in a computer, which is often
		>done, does not make the neural net a computer [...]and a        
		>suitable neural net might also be trained to emulate a 
		>computer [...] but that yet again does not show that neural 
		>nets are computers.
			  
I strongly disagree, and it has absolutely nothing to do with definitions. 
In general, if A and B are two information processing machines or systems 
of some sort, ANY sort, and A can simulate B, and B can simulate A, 
I find it impossible to avoid the conclusion that A and B are equivalent. 
For technological reasons we may find it more practical to build neural nets
rather than more conventional machines, just as we find it more practical to 
make computers out of semiconductors rather than vacuum tubes but that doesn't 
alter the fact that mathematically and philosophically they are equivalent.
			  
	      >the question of whether or not neural nets  are computers has
	      >at best a tangential relationship with the issue of  uploading.
	      
That I agree with.
	      

				   John K Clark           

-----BEGIN PGP SIGNATURE-----
Version: 2.6.i

iQCzAgUBL50m/303wfSpid95AQHJRATwguCqAkaBUk02s6ueSUDUTlunqGtMihA/
0IThPtH5v408dSyvdgl/ANbBE4HmWjbuMQvGqS/ZuRBhTyOzNC0dW4vHox4KEqgu
VEKgv96r57wXtANTU8VH08JghtaS+FtvzET63YejH0MGiPXisJ3hCEi1sw1Y9W4+
x/QJxqWT3CxlHHi0CMDQZcXI8D22p0tBIDBFpsMQwZrWMEe1lrw=
=cYAe
-----END PGP SIGNATURE-----


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=4289