Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!fas-news.harvard.edu!newspump.wustl.edu!news.ecn.bgu.edu!willis.cis.uab.edu!news.lsu.edu!darwin.sura.net!gatekeeper.es.dupont.com!esds01.es.dupont.com!slivova.es.dupont.com!owens
From: owens@slivova.es.dupont.com (Aaron J. Owens)
Subject: Short Course on Neural Networks at U. Del.
Message-ID: <1995Oct11.235146.6123@es.dupont.com>
Sender: news@es.dupont.com (USENET News System)
Nntp-Posting-Host: slivova.es.dupont.com
Organization: DuPont Experimental Station
X-Newsreader: TIN [version 1.2 PL0]
Date: Wed, 11 Oct 1995 23:51:46 GMT
Lines: 89


ENGINEERING OUTREACH at the UNIVERSITY OF DELAWARE presents a 
SHORT COURSE:

EMPIRICAL MODELING USING NEURAL NETWORKS 

Taught by Aaron J. Owens & Alicia M. Walsh
          DuPont Neural Network Resource Center 

February 5 & 6, 1996
University of Delaware Newark Campus
Pearson Hall, Room 116
Academy St. & Lovett Ave.
Newark, DE

Nonlinear empirical models using Artificial Neural Networks greatly
expand the capabilities of those in today's research and commercial
environments.  This hands-on short course provides technically
oriented professionals with the knowledge and skills needed to
develop these models applying what has been found to be the most
useful paradigm -- the back propagation network.  Rather than
focusing on historical and biological aspects, the course
concentrates on a specific methodology for using neural networks as
an empirical nonlinear regression tool in a wide variety of
practical applications.  You will actually develop models for
example datasets in application areas such as classification,
formulation, process modeling and control, and time series analysis.

WHO SHOULD ATTEND
This course is designed for technically oriented professionals in
all disciplines who have an interest in using neural networks to
develop empirical models to optimize the use of data in their
technical or business fields. Basic computer knowledge including
familiarity with PC's and Windows is needed. 

BENEFITS
This course will provide you with the depth of knowledge required to
apply artificial neural networks technology to your individual
technical area.  You are encouraged to bring data to be modeled (in
flat ASCII format on a 3.5" PC diskette) from your own application
area.  You will have the opportunity to: 
* focus on a sound methodology for applying the back propagation
   network; 
* learn to use the Neuro-Shell II software package on PC's; 
* model sample data sets and data from your own application field; 
* interact with others who utilize artificial intelligence
   techniques in their work. 

NOTE: You may purchase a take-home copy of Neuro-Shell II software 
at a significant discount along with your EARLY REGISTRATION for 
this short course.

THE AGENDA 
8:30 a.m. - Registration/Continental Breakfast 
The course will begin with a concept-based introduction to empirical
(data-driven) modeling and neural network definitions, theory and
principles. Using a format of lecture followed by workshops, the
basic steps of data analysis as they pertain  neural networks will
be discussed, covering the following issues: 
* amount and types of data needed;
* pre-processing of data;
* screening of data;
* using neural networks as an empirical nonlinear regression tool;
* model development using statistically-motivated methodology to
   obtain the optimal network structure; application areas will include
   classification, formulation, process modeling and control, and time
   series analysis; 
* summary example and workshop.

REGISTRATION:
The $795 program fee covers registration and handout materials,
continental breakfasts, lunches and all breaks.  PARTICIPANTS SHOULD
BRING THEIR OWN NEURO SHELL II SOFTWARE OR MAY PURCHASE THE NEURO
SHELL II SOFTWARE FOR AN ADDITIONAL $350 WHEN REGISTERING. (This 
software retails at $495.)

Deadlines:  If ordering software, register by January 15, 1996.
            If bringing own copy of software, register by January 
	    22, 1996.

TO REGISTER or FOR MORE INFORMATION, contact:
Kathy Werrell or Bob Sample
Engineering Outreach
University of Delaware
Newark, DE  19716-3101

phone:  302-831-2401
fax:    302-831-8179
email:  outreach@mvs.udel.edu
