Erik Hansen wrote:
took a course in Neural networking.
the prof said the computer code was too "dense"
and referred me to verbal passages that made no
sense
at all. in desperation i tried stepping through
the code
and it all came clear. the equations and graphics
were intuitive when seen one line of code at a
time .
Many years ago I had a subscription to AI Magazine, and they ran a good
many columns by a very talented writer on neural networks named Maureen
Caudill. She never lacked depth to her writings, but somehow managed
to explain things in ways that even a relative neophyte like me could
understand.
I've not seen much in the way of web-published work from her, but I'm
sure you could find some of her books through your local library.
In a quick Google for her name I stumbled across this item which may be
worth a look:
<http://web.informatik.uni-bonn.de/II/ag-klein/people/zach/benchmarks/NeuralNetTest.java>
Excerpt:
* Class NeuralNetTest
*
* Aka Back Propagationn Neural Net. This code is a modified
* version of the code that was submitted to BYTE Magazine by
* Maureen Caudill. It accompanied her article "Expert Networks,"
* BYTE, Oct. '91, though that article doesn't discuss this type
* of neural net algorithm in particular. For that, see "Back
* Propagation," BYTE, Oct. '97 by William P. Jones and Josiah
* Hoskins.
*
* The author's original heading/comment was as follows:
*
* Backpropagation Network
* Written by Maureen Caudill
* in Think C 4.0 on a Macintosh
*
* (c) Maureen Caudill 1988-1991
* This network will accept 5x7 input patterns and produce 8 bit
* output patterns. The source code may be copied or modified
* without restriction, but no fee may be charged for its use.
--
Richard Gaskin
Managing Editor, revJournal
_______________________________________________________
Rev tips, tutorials and more: http://www.revJournal.com
_______________________________________________
use-revolution mailing list
use-revolution@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution