X-Message-Number: 25627
From: 
Date: Wed, 26 Jan 2005 01:53:08 EST
Subject: Re: Limpinwood X-Prize: Butterflies vs. K-Spaces

From Peter Merel:
> 
> You assume
> 
> i) that artificial neurons are computationally equivalent to biological  
> neurons. This appears quite unlikely, at least in terms of current ANN  
> technologies.
> 
> ii) that I was correct in the number of neurons I quoted. While the  
> source I used seemed authoritiative, most of the articles I've found  
> online suggest 300,000 neurons is more like it. Mea culpa.
> 

i/ There are different "computer neurons"  Those of DeGaris for example looks 
very simplified. He choose them because he wanted as much of them as possible 
on few FPGAs. If you want more realistic neurons, far less can be implemented 
on a circuit.

ii/  Bee are said to have 800 000 neurons, so 300 000 may be right as an 

order of magnitude. On the other hand, 3 000 "good" (not perfect) neuron is what
can be currently implemented on the biggest FPGA. So I remain with this walue 

for my proposal : A 5 carats ruby for the drawing of such an ANN on a Spartan 3
FPGA.

Any taker ?

Yvan Bozzonetti.


 Content-Type: text/html; charset="US-ASCII"

[ AUTOMATICALLY SKIPPING HTML ENCODING! ] 

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=25627