[comp.ai.neural-nets] Neural network information wanted

richman@uxe.cso.uiuc.edu (07/15/88)

Could someone recommend a good introductory text which deals with
Neural Networks?  Please e-mail responses to:
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Mike Richman
uucp:  {ihnp4,seismo,puree,convex,uunet}!uiucdcs!uiucuxc!uiucuxe!richman
arpanet:  richman%uiucuxe@a.cs.uiuc.edu     bitnet: richman@uiucuxe
csnet:    richman%uiucuxe@uiuc.csnet        icbm:   40 05 N  /  88 14 W
internet: richman@uiucuxe.cso.uiuc.edu      milnet: richman@uiucuxe.arpa
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Thanks!

dtraver@macomw.ARPA (George Andrew Traver) (07/19/88)

In article <220700001@uxe.cso.uiuc.edu> richman@uxe.cso.uiuc.edu writes:
>
>Could someone recommend a good introductory text which deals with
>Neural Networks?  Please e-mail responses to:

Me too!  Email to dtraver@macomw.arpa.


Thanks in advance.



*******************************************************************************

Success has many fathers :-), while failure is an orphan. :-(

------------------------------------------------------------------------------

The views expressed in this article are shared by no other living being
or organization I know of. ( especialy my girlfreind )

	   		dtraver@macomw.arpa
	  
			FLAMES > /dev/null

*******************************************************************************

kennel@minnie.cognet.ucla.edu (Matthew Kennel) (07/19/88)

In article <348@macomw.ARPA> dtraver@macomw.UCSD.EDU (George Andrew Traver) writes:
>In article <220700001@uxe.cso.uiuc.edu> richman@uxe.cso.uiuc.edu writes:
>>
>>Could someone recommend a good introductory text which deals with
>>Neural Networks?  Please e-mail responses to:
>
>Me too!  Email to dtraver@macomw.arpa.

There is no comprehensive introductory textbook that I've seen that
deals with neural networks in anything more than a cursory treatment.

The standard reference is Rumelhart and McClelland's two volume work
"Parallel Distributed Processing.", published by the MIT press.

The first few chapters are not particularly
technical at all.  As a physics undergraduate, I found some of the
references to psychology works somewhat obscure, but the basic concepts and
mathematics were quite clear and simple.

Matt K.
(kennel@cognet.ucla.edu)

sbrunnoc@hawk.ulowell.edu (Sean Brunnock) (07/20/88)

From article <14531@shemp.CS.UCLA.EDU>, by kennel@minnie.cognet.ucla.edu (Matthew Kennel):
>>In article <220700001@uxe.cso.uiuc.edu> richman@uxe.cso.uiuc.edu writes:
>>>
>>>Could someone recommend a good introductory text which deals with
>>>Neural Networks?  Please e-mail responses to:
> 
> The standard reference is Rumelhart and McClelland's two volume work
> "Parallel Distributed Processing.", published by the MIT press.

I find the textbook "Explorations in Parallel Distributed Processing"
by the same authors to be a better book. It is more concise and contains
C source code for the IBM-PC.

		      Sean Brunnock

pi@pollux.usc.edu (Bill Pi) (07/20/88)

In article <14531@shemp.CS.UCLA.EDU> kennel@minnie.cognet.ucla.edu (Matthew Kennel) writes:
>In article <348@macomw.ARPA> dtraver@macomw.UCSD.EDU (George Andrew Traver) writes:
>>In article <220700001@uxe.cso.uiuc.edu> richman@uxe.cso.uiuc.edu writes:
>>>
>>>Could someone recommend a good introductory text which deals with
>>>Neural Networks?  Please e-mail responses to:
>>
>>Me too!  Email to dtraver@macomw.arpa.
>
>There is no comprehensive introductory textbook that I've seen that
>deals with neural networks in anything more than a cursory treatment.
>
>The standard reference is Rumelhart and McClelland's two volume work
>"Parallel Distributed Processing.", published by the MIT press.

Volume 3 of the series is also available now from MIT Press:
"Explorations in Parallel Distributed Processing: A Handbook of Models,
 Programs, and Exercises", J. L. McClelland and D. E. Rumelhart, 1988.
It comes with two 5.25" floopy disks, which contain a set of seven simulation
problems describe in the book.

Also, you might want to check the official Journal of the International
Neural Network Society (INNS) called "Neural Networks" from Pergamon
Journals, Inc.

>
>The first few chapters are not particularly
>technical at all.  As a physics undergraduate, I found some of the
>references to psychology works somewhat obscure, but the basic concepts and
>mathematics were quite clear and simple.

For people with physic's background, you might want to check out the following
articles and their references:

1. "Spin-glass models of Neural Networks", D. J. Amit, H. Gutfreund, and H.
     Sompolinsky, Physical Review A, Vol. 32, No. 2, 1985, pp. 1007-1018.

2. "Spin Glass Model for a Neural Network: Associative Memories stored with
     Unequal Weights", J. Phys. France (or J. de Physique), Vol. 49, 1988,
     pp. 167-174.

Also, I find

3. "Bidirectional Associative Memories", B. Kosko, IEEE Trans. on Sys. Man
and Cybern., vol. 18, No. 1, 1988, pp.49-60.

also interesting.

Greetings,
Jen-I Pi :-)			     UUCP:    {sdcrdcf,cit-cav}!oberon!durga!pi
Department of Electrical Engineering CSnet:   pi@usc-cse.csnet
University of Southern California    Bitnet:  pi@uscvaxq
Los Angeles, Ca. 90089-0781	     InterNet: pi%durga.usc.edu@oberon.USC.EDU

tyler@boulder.Colorado.EDU (Tyler Stevens) (07/28/88)

>There is no comprehensive introductory textbook that I've seen that
>deals with neural networks in anything more than a cursory treatment.
>Matt K.
>(kennel@cognet.ucla.edu)

I took a class with Paul Smolensky (of PDP-group fame), 
here at CU-Boulder - and I believe he is currently working on 
an introductory text to PDP, etc. (We used several
chapters of it; it is very good.) Keep your eyes pealed to this
space, and I'll post the name of it when it comes out sometime
in the indefinite future.

tyler
tyler@boulder.colorado.edu

demers@beowulf.ucsd.edu (David E Demers) (07/28/88)

Robert Hecht-Nielsen is writing a textbook on Neurocomputing.  It is
applications oriented; however, it outlines the history of the models
used and the field itself.  It also delves into the mathematics of
neural nets quite heavily, including problems to avoid in converting
data into an appropriate format for input to a net.  It will be
published by Addison-Wesley, due date is late 88.  It will be the
text from which he teaches his year-long course at UCSD.

Dave DeMers     demers@cs.ucsd.edu
UCSD C-014
La Jolla, CA