ANNIE'96 PRELIMINARY PROGRAM

ANNIE '96

ARTIFICIAL NEURAL NETWORKS IN ENGINEERING

Artificial Neural Networks, Fuzzy Logic and Evolutionary Programming for Designing Smart Engineering Systems

PRELIMINARY PROGRAM

November 10 - 13, 1996

Marriott Pavilion Hotel, St. Louis, Missouri

ORGANIZER

University of Missouri - Rolla

URL: http://www.umr.edu/~annie

AFFILIATED ORGANIZATIONS

Office of Naval Research, Arlington, VA

Dynamic Systems Control Division, ASME

Lockheed-Martin Corporation, Bethesda, MD

McDonnell Douglas Corporation, St. Louis, MO

American Institute of Aeronautics and Astronautics

Center for Optimization and Control, Washington University, St. Louis, MO

ABOUT THE ANNIE EMBLEM

From its rise in 1900 B.C., through its 800 years of existence, the Hittite civilization flourished as an empire and a culture in Mid-Anatolia (Present-day Turkey). This was the first civilization to incorporate hieroglyphics with "picture tablets" as a form of written communication. It is appropriate that the "ANNIE" emblem expresses this innovative form of communication as neural networks are a similar first step in the dialog between human and computers in the concept of learning. Behind the ANNIE tablet is a simple neural network. The "nodes" of the network are adorned with designs found in contemporary Mid-Anatolian ceramic pottery. The figures to the left of the network are Hittite hieroglyphics. These figures pass through the neural network and are translated into letters of the present-day phonetic alphabet: the linkage of old innovations to the new.

Charlie K. Dagli

ANNIE 1996

TIME
SUNDAY
MONDAY
TUESDAY
WEDNESDAY
8:30 - 9:00 a.m.

Opening Remarks and Welcome


9:00 - 10:00 a.m.

TUTORIALS

SA1.1 Wavelet Synthesis

Methods

Stephen W. Kercel

SA1.2 Adaptive Critics: Theory

and Applications

K. KrishnaKumar

SA1.3 AdvancedPattern

Recognition with Neural

Networks

Joydeep Ghosh

MA1.1 Plenary Session:

Neural Networks in Control:

Theory and Practice

Kumpati S. Narendra

TA1.1 Plenary Session:

Evolutionary Computation:

The New Synthesis

David Fogel

WA1.1 Plenary Session:

Why Function Approximation

Works So Well in Neural-Net

Computing and How to Make it

Work Even Better:

A Critique and a Perspective

Yo-Han Pao

10:30 - Noon


MA2.1 Learning Algorithms and Training I

MA2.2 Fuzzy Systems I (Fuzzy Logic)

MA2.3 Evolutionary

Programming I

MA2.4 Smart Engineering Systems I (Robotics)


TA2.1 Learning Algorithms

and Training IV

TA2.2 Fuzzy Systems IV

(Pattern Recognition)

TA2.3 Evolutionary

Programming IV

TA2.4 Smart Engineering

Systems IV (Novel

Applications)

WA2.1 Control I

WA2.2 Pattern Recognition III

WA2.3 Data Analysis

WA2.4 Smart Engineering

Systems VII (Prediction)

Noon - 1:30 p.m.


Luncheon Speaker:

A Unified Framework for Neural

Networks, Genetic Algorithms, and

Their Engineering Applications

Erol Gelenbe

Luncheon Speaker:

Fractal Behavior of the

Electrocardiogram: Distinguishing

Heart-Failure and Normal Patients

using Wavelet Analysis

Malvin C. Teich

Luncheon Speaker:

Fusion of Fuzzy and Chaos, and

its Application in Industrial Field:

Time Series Prediction

Tadashi Iokibe

1:30 - 3:00 p.m.

SP1.2 Tools for Discovering Patterns in Data

John F. Elder IV

SP1.3 Learning of Statistical Neural Networks and Their Applications

Okan Ersoy

SP1.4 Time-Frequency and Wavelet Methods with Engineering and Applications

Metin Akay

MP1.1 Learning Algorithms and Training II

MP1.2 Fuzzy Systems II (Control)

MP1.3 Evolutionary

Programming II

MP1.4 Smart Engineering Systems II ( Process

Monitoring)


TP1.1 Learning Algorithms

and Training V

TP1.2 Pattern Recognition I

TP1.3 Bio-Medical Engineering I

TP1.4 Smart Engineering

Systems V

(Manufacturing

Processes)

WP1.1 Control II

WP1.2 Pattern Recognition IV

WP1.3 Communications

WP1.4 Smart Engineering

Systems VIII

(Prediction)

3:30 - 5:00 p.m.

MP2.1 Learning Algorithms and

Training III

MP2.2 Fuzzy Systems III

(Control)

MP2.3 Evolutionary

Programming III

MP 2.4 Smart Engineering

Systems III (Process

Monitoring)

TP2.1 Learning Algorithms and

Training VI

TP2.2 Pattern Recognition II

TP2.3 Bio-Medical

Engineering II

TP2.4 Smart Engineering

Systems VI (General

Applications)

WP2.1 Control III

WP2.2 Optimization

WP2.3 Decision Analysis

WP2.4 Smart Engineering

Systems IX (Prediction)

6:00 - 8:00 p.m.
RECEPTION



7:00 BANQUET and AWARD

PRESENTATION

Intelligent Control in Practice

Rahmat Shoureshi



WELCOME

On behalf of the organizing committee I would like to invite you to attend ANNIE '96, an international conference to be held on November 10-13, 1996, at Marriott's Pavilion Hotel in downtown, St. Louis, Missouri. This will be the sixth international gathering of researchers interested in engineering applications of neural networks, fuzzy logic and evolutionary programming. The previous conferences drew approximately 150 papers each from approximately twenty countries and their proceedings were published by ASME Press as hardbound books in five volumes titled Intelligent Engineering Systems Through Artificial Neural Networks, Fuzzy Logic and Evolutionary Programming, edited by Dagli, C.H., et. al.

The conference will cover the theory of artificial neural networks, fuzzy logic, and evolutionary programming, and their applications in the engineering domain. The objective of the conference is to bring together researchers in the field of neural networks, fuzzy logic, and evolutionary programming, and exchange ideas on their applications to engineering problems. Presentations dealing with applications of these emerging technologies are encouraged in the areas of: manufacturing engineering, biology and medicine, pattern recognition, image processing, process monitoring, control, recent theoretical developments in neural network architectures, fuzzy logic, evolutionary programming, fractals, chaos, and wavelets that can impact engineering applications.

The response to ANNIE '96 was excellent with over 250 abstracts received. Papers submitted based on these abstracts were reviewed by two referees and all accepted papers are included in the conference proceedings to be published by ASME Press as a hardbound book titled Smart Engineering System Design: Neural Networks, Fuzzy Logic, and Evolutionary Programming, edited by Drs. Dagli, Akay, Chen, Fernández, and Ghosh.

There are seven plenary sessions scheduled for ANNIE '96. Dr. Bob Narendra will speak on "Neural Networks for Control: Theory and Practice", during Monday plenary session. At noon, "A Unified Framework for Neural Networks, Genetic Algorithms, and Their Engineering Applications" will be presented by Dr. Erol Gelenbe. In Tuesday's plenary session, "Evolutionary Computation: The New Synthesis" will be introduced by Dr. David Fogel. Dr. Malvin Teich will cover "Fractal Behavior of the Electrocardiogram: Distinguishing Heart-Failure and Normal Patients Using Wavelet Analysis," in his luncheon address on Tuesday. Our honored banquet speaker for ANNIE '96 is Dr. Rahmat Shoureshi, from the Colorado School of Mines, who will discuss the topic "Intelligent Control in Practice." "Why Function Approximation Works so Well in Neural-Net Computing and How to Make it Work Even Better" will be the title of Dr. Yo-Han Pao's morning plenary session on Wednesday. For the luncheon address for Wednesday, Dr. Tadashi Iokibe will present his talk on "Fusion of Fuzzy and Chaos, and its Application in Industrial Field: Time Series Prediction."

In addition, half-day tutorials have been scheduled for Sunday, November 10th. These state-of-the-art workshops will cover the following areas:

Wavelet Synthesis Methods

Adaptive Critics: Theory and Applications

Advanced Pattern Recognition with Neural Networks

Tools for Discovering Patterns in Data

Learning of Statistical Neural Networks and Their Applications

Time-Frequency and Wavelet Methods with Engineering and Applications

I would like to thank members of the Organizing Committee, Sponsoring Organizations, and Co-Chairs, Drs. Akay, Chen, Fernández, and Ghosh for putting together an excellent program for ANNIE '96. I would like to recognize the excellent and timely efforts of the referees and contributors which made the conference possible.

We are looking forward to your visit and participation at the meeting.

Sincerely,

Cihan H. Dagli

Conference Chairman

ORGANIZING COMMITTEE

Conference Chair:

Prof. Cihan H. Dagli, Engineering Management, University of Missouri-Rolla, Rolla, MO

Conference Co-Chairs:

Biology and Medicine: Prof. Metin Akay, Biomedical Engineering, Rutgers University, Piscataway, NJ

Applications: Prof. C.L. Philip Chen, Computer Science and Engineering, Wright State University, Dayton, OH

Control: Prof. Benito Fernández, Mechanical Engineering, University of Texas at Austin, Austin, TX

Pattern Recognition: Prof. Joydeep Ghosh, Electrical and Computer Engineering, University of Texas at Austin, Austin, TX

Members:

USA

Prof. Luke Achenie, University of Connecticut

Prof. Hojjat Adeli, Ohio State University at Columbus

Prof. In Soo Ahn, Bradley University

Prof. Marijke F. Augusteijn, University of Colorado

Prof. Nirmal Bose, Pennsylvania State University

Prof. Anna L. Buczak, AlliedSignal, Inc.

Prof. Laura I. Burke, Lehigh University

Prof. Thomas P. Caudell, University of New Mexico

Dr. Tien-Hsin Chao, Jet Propulsion Lab

Prof. Rama Chellappa, University of Maryland

Prof. John Y. Cheung, University of Oklahoma

Prof. Maurice E. Cohen, California State University

Prof. Yechiel Crispin, Embry-Riddle Aeronautical University

Dr. Joel Davis, ONR

Prof. Mark Embrechts, Rensselaer Polytech. Inst.

Prof. Okan Ersoy, Purdue University

Prof. Laurene Fausett, Florida Institute of Technology

Prof. Jim Garrett, Carnegie-Mellon University

Prof. Jamshid Ghaboussi, University of Illinois-Urbana

Prof. David Goldberg, University of Illinois

Prof. Bruce Golden, University of Maryland

Prof. Sam Haddad, Santa Clara University

Prof. Thomas L. Hemminger, Penn State University

Prof. Michael Jordan, MIT

Prof. Sagar V. Kamarthi, Northeastern University

Prof. Nicolaos Karayiannis, University of Houston

Prof. R.L. Kashyap, Purdue University

Dr. Fatih Kinoglu, 3M Corporation

Prof. Soundar R.T. Kumara, Penn State University

Prof. Peter F. Lichtenwalner, McDonnell Douglas Aerospace

Prof. Chun-Shin Lin, University of Missouri-Columbia

Dr. Kenneth Marko, Ford Motor Company

Prof. Andrew J. Meade, Rice University

Prof. Alexander M. Meystel, Drexel University

Prof. Evangelia Micheli-Tzanakou, Rutgers University

Prof. Haluk Ogmen, University of Houston

Prof. Tokunbo Ogunfunmi, Santa Clara University

Prof. Özcan Özdamar, University of Miami-Coral Gables

Prof. James Peterson, Clemson University

Prof. Ervin Rodin, Washington University-St. Louis

Prof. Steven K. Roger, Air Force Inst. of Technology

Prof. Ryan G. Rosandich, University of Kansas

Prof. V. David Sanchez, University of Miami

Prof. Mohammed Sayeh, SIU-Carbondale

Prof. Ishwar K. Sethi, Wayne State University

Prof. Yung C. Shin, Purdue University

Mr. Patrick K. Simpson, Scientific Fishery Systems, Inc.

Prof. Alice E. Smith, University of Pittsburgh

Dr. Robert S. Smithson, Lockheed Missiles & Space Co.,Inc.

Dr. Dejan Sobajic, Electric Power Res. Inst.

Dr. Donald Specht, Lockheed Missiles and Space Co., Inc.

Prof. Daniel St. Clair, University of Missouri-Rolla

Prof. John P.H. Steele, Colorado School of Mines

Dr. Ted H.-T. Su, Honeywell Inc.

Dr. Bobby Sumpter, Oak Ridge National Laboratory

Dr. Richard Sutton, GTE Labs MS-44

Prof. Harold Szu, NSWC

Prof. Janet M. Twomey, Wichita State University

Prof. Hugh F. VanLandingham, Virginia Polytechnic

Dr. Paul Werbos, National Science Foundation

Prof. Darrell Whitley, Colorado State University

Prof. Yuehwern Yih, Purdue University

Prof. Francis T. Yu, Pennsylvania State University

Prof. Jacek M. Zurada, University of Louisville

CANADA

Prof. Michael Guillot, University of Laval, Quebec

Prof. Edwin P. Nowicki, University of Calgary

Mr. John Sutherland, AND America Ltd.

Prof. R.W. Toogood, University of Louisville

JAPAN

Prof. Yasuhiko Dote, Muroran Institute of Technology

Prof. Yukinori Kakazu, Hokkaido University

Prof. Masaharu Kitamura, Tohoku University

Prof. Kazuhiro Ohkura, Kobe University

Prof. Setsuo Ohsuga, University of Tokyo

Prof. Youchi Okabe, University of Tokyo

UNITED KINGDOM

Prof. J.N. Carter, Imperial College, U.K.

Prof. T. Sezgin Daltaban, Imperial College, England

Prof. Muhammed A. Javed, Southampton Inst., England

FRANCE

Prof. Stephane Canu, Universite de Compiègne, Cédex

Prof. Thierry Denoeux, Universite de Compiègne, Cédex

Prof. Bernard Dubuission, University of Compiègne, Cédex

Dr. Françoise Fogelman Soulie, Sligos

BELGIUM

Prof. Marc Acheroy, Royal Military Academy

Prof. Wim Mees, Royal Military Academy

Prof. F. Vandamme, BIKIT, University of Ghent

GERMANY

Prof. Winfried Schauer, Tech Hoch. Schule Wismar, GERMANY

Prof. Paulo Camargo Silva, University of Erlangen-Nurnberg

SWITZERLAND

Prof. Ralf Salomon, University of Zürich

AUSTRALIA

Prof. K.C. Chan, University of New South Wales

Prof. T.S. Dillon, La Trobe University

Prof. D.J. Gunaratnam, University of Sydney

ITALY

Prof. G. Di Stefano, Universita degli Studi de L'Aquila

ST. LOUIS ATTRACTIONS

Diversity and expansion have been St. Louis hallmarks since its founding. The area offers a wide variety of attractions, both indoor and out, to please almost everyone. The Gateway Arch, located on the riverfront, marks the starting point of America's westward expansion in the 1800's. Also on the riverfront is Laclede's Landing, where 19th century warehouses have become a collection of offices, restaurants, shops, galleries and lively lounges, complete with cast iron street lamps and cobblestone streets. Sports fans may want to visit Busch Stadium, or the St. Louis Sports Hall of Fame located inside. St. Louis has a wealth of fine restaurants, with a wide variety of ethnic and regional cuisine. For shopping, you may choose from St. Louis Centre, Union Station - a renovated Romanesque-style train terminal which features boutiques and other unique shops and restaurants, Plaza Frontenec, Westport Plaza and a pleasing variety of others.

St. Louis offers many other exciting attractions, events and activities for the entire family, many of which are free. St. Louis Visitor's Brochures will be enclosed in each participant's conference packet upon entrance to the conference. For more information about St. Louis, or for your personal copy of the official St. Louis Visitor's Guide, at no cost, please call the St. Louis Convention and Visitors Commission at 1-800-247-9791, or local at 314-421-1023, Fax 314-421-0039. A copy of the official guide will be on-hand at the conference table for your convenience. You are welcome to visit the World Wide Web site of at http://www.st-louis.mo.us.

CONFERENCE SITE

All activities of the Conference will be held at the Marriott Pavilion Downtown, One Broadway Street, St. Louis, MO. See HOTEL INFORMATION (next page) for further details on room rates and facilities provided by the Marriott.

The Marriott is located in downtown St. Louis and is convenient to many area attractions, such as the Gateway Memorial Arch, Old Courthouse, Busch Stadium, Laclede's Landing, Anheuser-Busch Brewery, and St. Louis Centre.

If you desire, there is an Airport Express Shuttle to the Hotel at a charge of $10.00 one way, or $15.00 round trip. Major car rentals and taxis will be available from the airport. Taxi fares from the Lambert International Airport range from $18.00 - $20.00. The Metro-link light rail runs from the Lambert International Airport to Busch Stadium (two blocks from Marriott-downtown) for a fare of $1.00.

If you should need assistance with making your reservations, please contact Rita Schneider at (573) 341-6576, or through e-mail address: rita@shuttle.cc.umr.edu. Please send all hotel reservations to the Marriott, NOT to the University.

SOCIAL FUNCTIONS

RECEPTION AND BANQUET

All registrants are invited to attend a reception on Sunday, November 10, 1996 from 6:00 - 8:00 p.m. at the Marriott. The conference banquet will be held at 7:00 p.m., Tuesday, November 12th. Banquet tickets included in the registration package must be brought to the banquet. Extra tickets are available for guests at a cost of $34 each.

COFFEE BREAKS

Coffee and soft drinks will be made available during breaks in the morning and afternoon near the registration area at the following times: 10:00 - 10:30 a.m. and 3:00 - 3:30 p.m.

SPEAKERS BREAKFAST

A complementary continental breakfast will be available from 7:30 a.m . to 8:30 a.m. on November 11-13 for Speakers and Session Chairs of the day. Speakers will have an opportunity to meet with their fellow speakers and session chairs at tables marked with their session number. The "Chairs" will have an opportunity to meet with speakers to share logistical information related to the session activities. All speakers are urged to attend breakfast on the day of their session and meet the chairs of their session.

CONFERENCE PROCEEDINGS

A copy of the conference proceedings, published by ASME press in a hardbound book, titled Smart Engineering System Design: Neural Networks, Fuzzy Logic and Evolutionary Programming, edited by Drs. Dagli, Akay, Chen, Fernández, and Ghosh, will be provided to all registrants at the conference. After the conference, the book will be available from the publisher.

HOTEL INFORMATION

The Marriott Pavilion Downtown is a new hotel with modern and well designed facilities for holding conferences. It has 670 deluxe guest rooms, executive, and hospitality suites. Amenities include indoor pool, whirlpool, sauna, health club and weight room, and an attached parking garage for your convenience. A special conference rate of $82.00 per night, for both single and double occupancy is available to attendees of the ANNIE '96 conference up until the cut-off date of October 21, 1996. To reserve your room, please send the enclosed reservation form and a first night's deposit to:

Marriott Pavilion Hotel

One Broadway St.

St. Louis, MO 63102-1772

(314) 421-1776

The Marriott Pavilion is located on the corner of Market & Broadway, overlooking Busch Stadium, 2 blocks from the Gateway Arch & the Mississippi riverfront and within easy walking distance of downtown St. Louis' shopping and business district.

If you are driving

Westbound: I-55 to Memorial Drive Exit Turn left at the second traffic signal (Market St.) Go 2 blocks to Broadway & turn left.

Eastbound: Via or from Lambert Intl. Airport. I-70 to Memorial Drive Exit Go to 3rd traffic signal (Market St.) & turn right Go 2 blocks to Broadway & turn left.

Northbound: I-55 to I-44 exit at Memorial Drive turn left at the second traffic signal (Market St.) Go 2 blocks to Broadway & turn left.

CONFERENCE REGISTRATION

Advance registration and payment of fees as shown below is strongly encouraged and will be appreciated. This helps the conference staff in minimizing the time that attendees have to spend at the registration desk and improves efficiency. Please note that the cutoff date for receipt of fees is October 25, 1996. Full payment in U.S. Currency by money order, bank draft, or check must accompany all advance registration. The registration schedule is as follows:

ADVANCE CONFERENCE REGISTRATION

Registration Fee $410.00

(Includes reception, banquet, three luncheons and one copy of the proceedings).

Tutorials $110.00

Extra Banquet Ticket (Spouse or Guest) $ 34.00

Extra Luncheon (Each day) $ 23.00

ON SITE REGISTRATION

Registration Fee $430.00

(Includes reception, banquet, three luncheons and one copy of the proceedings).

Tutorials $110.00

Extra Banquet Ticket (Spouse or Guest) $ 34.00

Extra Luncheon (Each day) $ 23.00

REGISTRATION FORM AND TIMES

Please use the Conference Registration Form, found at the end of this booklet, and mail with due amount of fees to:

ANNIE '96

c/o Rita Schneider

236 Smart Engineering Systems Lab

Dept. of Engineering Management

University of Missouri-Rolla

1870 Miner Circle

Rolla, MO 65409-0370, USA

Phone: 573-341-6576

FAX: 573-341-6567

E-mail: rita@shuttle.cc.umr.edu

URL: http://www.umr.edu/~annie

The conference registration desk will be open during the following hours:

Sunday, November 10, 1996 8:00 a.m. - 6:00 p.m.

Monday, November 11, 1996 8:00 a.m. - 6:00 p.m.

Tuesday, November 12, 1996 8:00 a.m. - 6:00 p.m.

Wednesday, November 13, 1996 8:00 a.m. - 5:00 p.m.

KEY TO SESSION AND TUTORIAL ABBREVIATIONS

The Tutorials and Sessions are noted by Day, Time of Day, Session Order and Room Number.

Example: "MA1.1"

M - Day of the Week (Sunday, Monday, Tuesday, Wednesday)

A - Morning or Afternoon Session (A.M. or P.M.)

1. - First or Second Session (1, or 2)

1 - Room Designation (1, 2, 3, 4)

TUTORIALS

The first day of the conference, Sunday, November 10, 1996, is allocated for half-day tutorials in various applications of Artificial Neural Networks in Engineering. They are:

Sunday, 9:00 - Noon

SA1.1 Wavelet Synthesis Methods

Dr. Stephen W. Kercel, Oak Ridge National Lab, Oak Ridge, TN

SA1.2 Adaptive Critics: Theory and Applications

Dr. K. KrishnaKumar, The University of Alabama, Tuscaloosa, AL

SA1.3 Advanced Pattern Recognition with Neural Networks

Dr. Joydeep Ghosh, University of Texas at Austin, Austin, TX

Sunday, 1:30 - 4:30 p.m.

SP1.2 Tools for Discovering Patterns in Data

Dr. John F. Elder IV, University of Virginia, Charlottesville, VA

SP1.3 Learning of Statistical Neural Networks and Their Applications

Dr. Okan Ersoy, Purdue University, West Lafayette, IN

SP1.4 Time-Frequency and Wavelet Methods with Engineering and Applications

Dr. Metin Akay, Rutgers University, Piscataway, NJ

Sunday, 9:00 - Noon

SA1.1 - Wavelet Synthesis Methods

Dr. Stephen W. Kercel, Oak Ridge National Lab, Oak Ridge, TN

Scope: Wavelet Analysis is one of the most explosively growing new developments in signal processing. The continuous wavelet transform was introduced in 1984 by French geologists as a method of identifying the signatures of mineral deposits in geophysical data. In 1986 the perfect reconstruction subband coder was developed for data compression in digital telephone systems. Shortly thereafter, the relationship between the discrete wavelet transform and the subband coder was discovered. This led to a fast implementation of the wavelet transform. With the derivation of a practical orthonormal wavelet by Daubechies in the late eighties, the way was cleared for a wide range of engineering applications. Wavelet methods closely emulate the processes at work in biological sensory systems (such as eyes) that feed information to the brain; not surprisingly, wavelet analysis provides a powerful method for conditioning data for presentation to neural networks. Wavelet methods are used for data compression, pattern recognition, image enhancement, denoising, and a variety of other applications. Dedicated hardware, with embedded wavelet processes, is beginning to appear.

Overview: Wavelet concepts are becoming reasonably well known. What remains somewhat obscure are the methods for synthesizing practical systems based on wavelets. There are several powerful synthesis methods that give the engineer the ability to design "made to order" wavelet systems. The fast wavelets transform, and it's, inverse are implemented using a paraunitary subband coder; the algorithm is much faster and cheaper than the more widely known fast Fourier transform. Any filter bank that meets the paraunitary constraints can be represented as a cascaded lattice, and this gives rise to a method, based on optimization of the lattice coefficients, to synthesize any kind of discrete wavelet filter or perfect reconstruction multiresolution filter. Alternatively, the power spectrum of a wide range of wavelet filters can be synthesized by application of the Bernstein polynomial. Given the power spectrum, the wavelet filter can be derived by spectral factorization. There are a number of factorization algorithms, one of the less inconvenient being based on the complex spectrum. Practical wavelet functions are not generally expressible in closed form; instead they are defined as a recursive function of their filter coefficients. The structure of the recursion leads to the cascade algorithm, an elegant, and fast, method for plotting the wavelet function implemented by a given filter. The basic concepts behind these methods as well as their limitations and implementations are presented in this tutorial.

Instructor's Background: Dr. Kercel is a Development Engineer in the Instrumentation and Controls Division of Oak Ridge National Laboratory, and an Adjunct Assistant Professor of Electrical Engineering at the University of Tennessee. He won the 1995 R&D 100 Award for application of wavelets to the development of an instrument to monitor magnetic field effects in nuclear power plant control rooms. He is a judge in the 1996 R&D 100 competition. He has published several papers on the application of wavelets to pattern recognition, and the implementation of wavelets in dedicated real-time hardware. His research interests are anticipatory systems, safety and security systems, wavelet hardware implementations, electromagnetic interference, optical systems, and pattern recognition. Dr. Kercel is a Senior Member of IEEE, a member of Eta Kappa Nu, Tau Beta Pi, Phi Kappa Phi, and Toastmasters International. He is a registered Professional Engineer of Tennessee.

Sunday, 9:00 - Noon

SA1.2 - Adaptive Critics: Theory and Applications

Dr. K. KrishnaKumar, The University of Alabama, Tuscaloosa, AL

Scope: This tutorial takes a hard look at the methods and benefits of using critics that not only predict future performances but also adapt themselves. Adaptive critics have been tested and shown to be superior approaches for adaptive control problems. Some of the recent applications include, aircraft landing control, aircraft control through turbulence, automotive control, and inventory control. Adaptive critic methods take us a step closer to true intelligent control as they provide three degrees of freedom (cost identification, system identification, and control adaptation) as opposed to the traditional two-degree-of freedom (system identification and control adaptation) adaptive controllers. The primary goal of this tutorial is to ensure that every participant is able to understand the theory and the application potential of adaptive critics. An intelligent control complexity test bed software will be used to illustrate the benefits of adaptive critics. This PC-based software will be given free to the attendees.

Overview: A critic is one that differentiates between good and bad performances. Critics vary from simple critics that give binary outputs to more sophisticated critics that not only predict future performances but also adapt to the changing operating environments. Sophisticated critics in general follow the principles of dynamic programming. In this tutorial we will examine: simple critics; adaptive critics based on heuristic dynamic programming, dual heuristic dynamic programming, and combinations; and immunized critics. The tutorial is divided into four sessions (a) Neuro-controller primer, (b) Dynamic programming primer, (c) Critics and Adaptive Critics; and (d) Applications.

Instructor's Background: Dr. KrishnaKumar is an Associate Professor of Aerospace Engineering and the Director of the Intelligent Control Laboratory at the University of Alabama, Tuscaloosa. His externally funded research include immunized artificial neural systems; neuro-control and neural network applications to OPAD, automated training, aircraft and spacecraft control, inventory control, and large space structures; Evolutionary Fuzzy Modeling (EFM) applications to F-18 stability augmentation systems, helicopter hover training, genetic algorithm applications to flight control, adaptive control, and partitioned controllers, and flight reconstruction of recorded flights using robust estimators. He is the chairman of the AIAA artificial intelligence technical committee, member of the SAE simulation technology committee, associate editor of the AIAA Journal of Guidance, Control, and Dynamics, and a member of the Journal of Aircraft editorial board. He has organized a number of tutorial workshops and short courses on genetic algorithms, fuzzy logic, and neural networks.

Sunday, 9:00 - Noon

SA1.3 - Advanced Pattern Recognition with Neural Networks

Dr. Joydeep Ghosh, University of Texas at Austin, Austin, TX

Scope: The effective use of neural network technology for pattern recognition involves not only an understanding of the algorithms involved, but also data selection and presentation issues as well as the interpretation and analysis of results. We will consider several "neural network" approaches and put them in perspective with respect to more established methods. Advanced techniques such as classifier combining/hybridization, inverse problem modeling and estimation of fundamental limits such as Bayes error, will also be discussed. Thus, the tutorial is aimed at engineers and scientists who want to further enhance their understanding of pattern recognition and its applications.

Overview: Over the past forty years, a solid theory of statistical pattern recognition has been developed in response to the vast practical applications of pattern recognition. Some of the recent neural network techniques are very similar to classical methods, while others provide new capabilities. The talk will examine the pros and cons of these newer methods as compared to classical techniques, based both on theoretical concepts as well as recent experimental results. Links between neural networks and statistical techniques will be drawn and practical implications will be highlighted. Some recent techniques that deal with difficult pattern recognition problems will then be presented and critically evaluated.

Instructors' Background: Joydeep Ghosh is an Associate Professor in the Department of Electrical and Computer Engineering at the University of Texas, Austin, where he holds the Endowed Engineering Foundation Fellowship.

Dr. Ghosh teaches graduate courses on Artificial Neural Systems and Neural Networks for Pattern Recognition. He has more than seventy refereed publications in these areas and has applied neural network technology to several industrial problems including manufacturing process modeling, medical analysis, image analysis and sonar classification. Dr. Ghosh received the 1992 Darlington Award for best journal paper from IEEE Circuits and Systems Society, and also awards for four papers on neural network theory and applications. He is an associate editor of "IEEE Trans. Neural Networks" and of "Pattern Recognition."

Sunday, 1:30 - 4:30 p.m.

SP1.2 - Tools for Discovering Patterns in Data

Dr. John F. Elder IV, University of Virginia, Charlottesville, VA

Scope: This tutorial surveys the leading inductive algorithms for estimation and classification employed in industry and academia. We will examine the key inner workings of competing methods, compare their merits, and view their effectiveness on practical applications. We will first review classical statistical techniques, both linear and nonparametric, then outline the ways in which these basic tools are modified and combined into more modern methods, from smoothing kernels to radial basis functions. We'll focus on three popular approaches: neural networks, polynomial networks, and decision trees, using example data from recent scientific, medical, and financial applications to demonstrate effective techniques employed by experienced analysts (including feature extraction, scientific visualization, resampling, and "model fusion").

Overview: Increasingly powerful techniques for "mining" useful information from vast datasets are emerging from diverse disciplines. This tutorial provides a general framework for the field as well as distinctives for each of the leading algorithms. It emphasizes understanding of the key issues in inductive modeling and classification, and aims to reveal the inner workings of representative tools. Practical techniques, including recent methods for combining and enhancing the algorithms, are emphasized over theoretical results.

Instructor's Background: John Elder has more than a decade of experience in applying adaptive, data-driven techniques to practical problems, and has developed or refined some of the methods covered in this course. He is an independent consultant to industry and Adjunct Professor at the University of Virginia, and has authored four book chapters and numerous articles on adaptive methods of pattern discovery. He has been a researcher at Rice University and at an engineering consulting firm, and was Director of Research for an investment management company. Dr. Elder is a frequent lecturer on pattern discovery techniques, and is the technical chair of the Adaptive and Learning Systems Group of the IEEE Systems, Man, and Cybernetics Society.

Sunday, 1:30 - 4:30 p.m.

SP1.3 - Learning of Statistical Neural Networks and Their Applications

Dr. Okan Ersoy, Purdue University, West Lafayette, IN

Scope: Learning techniques are essential to the success of neural networks in practical applications. A number of such techniques are directly or indirectly related to probability and statistics. Understanding of the underlying issues is crucial to the effective use of neural networks in applications, as well as for developing new learning algorithms. This tutorial will focus upon these relationships and give examples of their uses in major applications such as nonlinear prediction, system identification, recognition, and classification.

Overview: Most popular ANN architectures in practice are multistage neural networks such as backpropagation. Their learning is often based on minimizing the square error at the output. If the output representation is chosen correctly, they actually estimate the a posteriori class probabilities. How this can be done in better ways so that neural networks actually perform better than classical techniques is one topic of the tutorial. Some new learning algorithms use Bayesian techniques to learn the weights of the networks. Such techniques are expected to generalize better during testing. Their advantages and disadvantages will be described. Another connection to statistics is how the weights of the network are distributed and how they can be pruned by using statistical rules. There are many other examples of how neural networks and statistical techniques as well as their applications are intertwined. The tutorial will highlight these interrelationships in detail.

Instructors' Background: Okan K. Ersoy was born in Istanbul, Turkey, on September 5, 1945. He received the B.S.E.E. degree from Robert College in 1967, and the M.S., Certificate of Engineering, second M.S., and Ph.D. degrees from the University of California, Los Angeles, in 1968, 1971, and 1972, respectively.

He was a Teaching and Research Assistant in the Department of Electrical Sciences and Engineering, UCLA (1968-1972), Assistant Professor in the Department of Electrical Engineering, Bosphorus University (1972-1973), and Associate Professor in the second semesters at the same university (1976-1980). He joined the Center for Industrial Research, Oslo, Norway, as a Researcher in the Computer Science Division in 1973. He was a Visiting Scientist at UCSD In 1980-1981. He has been with Purdue University, School of Electrical and Computer Engineering, West Lafayette, Indiana since August 1985.

Dr. Ersoy was a Fullbright Fellow in 1967-68. His current interest include neural networks and applications; digital signal/image, processing and recognition, fast algorithms and architectures, optical information processing, diffractive optics and holography.

He has published approximately 150 papers in his areas of interest. He also holds three patents which are separately patented in USA, Norway, Denmark, and Sweden. He has been a referee for IEEE Tran. Acoustics, Speech, Signal Processing, IEEE Tran. Computer, Applied Optics, Optical Engineering, a number of other journals, and granting agencies such as National Science Foundation. He is associate editor of IEEE Tran. Neural Networks, IEEE Tran. Circuits and Systems and International Journal of Smart Engineering Design. He has worked with numerous projects in his areas of interest in USA, Norway, and Turkey.

Sunday, 1:30 - 4:30 p.m.

SP1.4 - Time-Frequency and Wavelet Methods with Engineering and Applications

Dr. Metin Akay, Rutgers University, Piscataway, NJ

Scope: A signal can be considered to be stationary if its statistical characteristics are NOT changing with time. Stationary signals can be analyzed using classical Fourier transform methods in which the signal can be expanded on the orthogonal basis functions (sin and cosine waves). However, most biomedical signals are nonstationary and have highly complex time-frequency characteristics. In practice, the stationary condition for the nonstationary signals can be satisfied by dividing the signal into blocks of short segments in which the signal segment can be assumed to be stationary.

This method, called the short time Fourier transform (STFT) was proposed by Gabor in 1946. However, the problem with the STFT is the length of the desired segment. Choosing a short analysis window may cause poor frequency resolution. On the hand, a long analysis window may improve the frequency resolution but compromises the assumption of stationarity within the window. To overcome these difficulties with the STFT, several time-frequency analysis methods including the Gabor representation, Wigner-Ville Distribution, Binomial transform, Choi-Williams, Reduced Interference Distribution methods etc. have been proposed. All the time-frequency methods have been unified by Cohen.

An alternative way to analyze the nonstationary biomedical signals is the wavelet transform which expands the signal onto the basis functions. The basis functions can be constructed by dilation, contractions and shifts of a unique function called the wavelet prototype. The wavelet method acts as a mathematical microscope in which we can observe different parts of the signal by just adjusting the focus.

Overview: Time-frequency and Time-scale analysis methods have been widely used in the signal processing of signals. These methods represent the temporal characteristics of a signal by its spectral components in the frequency domain. In this way, important features of the signal can be perceived and analyzed in order to understand or model the physiological system. In this presentation, we will review some useful analysis methods, such as the short-Time Fourier Transform, the Gabor Representation, the Wigner-Ville Distribution, the Exponential Distribution, and the Wavelet Transforms. The basic concepts behind these methods as well as their limitations, implementation, and medical applications are presented.

Instructor's Background: Dr. Metin Akay is currently a visiting professor at Rutgers University. He is co-author with Dr. Welkowitz and Dr. Deutch for the new edition of the book entitled Theory and Design of Biomedical Instruments (Academic Press, 1991) and is the author of Biomedical Signal Processing (Academic Press, 1993) and Detection and Estimation of Biomedical Signals (Academic Press 1996). He is also the author of the book entitled Time-Frequency and Wavelets in Biomedical Engineering (IEEE Press, 1996). He was the guest editor of the Engineering (IEEE EMB Magazine on the Fuzzy Logic in Medicine and Biology and on the Time-Frequency and Wavelet Analysis of Biomedical Signals). He is a Co-chair of the International conference on the Artificial Neural Networks in Engineering (ANNIE'95, and ANNIE'96).

Dr. Metin Akay's research areas of interest are fuzzy neural networks, biomedical signal processing, wavelet transform, detection and estimation theory and application to biomedical signals. His biomedical research areas include maturation, breathing control, noninvasive detection of coronary artery disease, and the understanding of the autonomic nervous system. Dr. Akay is a senior member of IEEE, a member of ETA Kappa, Sigma Xi, Tau Beta Pi, the American Heart Association, and the New York Academy of Science.

PLENARY SPEAKERS

Monday, November 11, 9:00 - 10:00 a.m.

Neural Networks for Control: Theory and Practice

Kumpati S. Narendra, Yale University, New Haven, CT

Kumpati S. Narendra received the Ph.D. degree from Harvard University, Cambridge, MA, in 1959. At present, he is Professor of Electrical Engineering and Director of the Center for Systems Science at Yale University. He is the co-author of three books, "Frequency Domain Criteria for Absolute Stability" with J. H. Taylor "Stable Adaptive Systems" with A. M. Anaswamy; and "Learning Automata--An Introduction" with M. A. L. Thatachar. He is also the editor of four books. His research interests include stability theory, adaptive systems, learning automata, and the control of complex systems using neural networks.

Dr. Narendra is a member of Sigma Xi, and a Fellow of the American Association for the Advancement of Science and IEEE. He was the recipient of the 1972 Franklin V. Taylor Award of the IEEE systems, Man, and Cybernetics Society, and the George S. Axelby Best Paper Award of the IEEE Control Systems Society in 1988, the Education Award of the American Automatic Control Council in 1990, and the Outstanding Paper Award of the Neural Networks Council in 1991. In 1994, he received the Neural Networks Leadership Award of the International Neural Networks Society. He was also appointed Distinguished Visiting Scientist by the Jet Propulsion Laboratory for the year 1994-1995. He received the honorary D. Sc. degree from Anna University, Madras, India and the Bode Prize of the Control Systems Society in 1995. He delivered the Bode Lecture at the 1996 Conference on Decision and Control in New Orleans.

Tuesday, November 12, 9:00 - 10:00 a.m.

Evolutionary Computation: The New Synthesis

David Fogel, Natural Selection, Inc., La Jolla, CA

Dr. Fogel is executive vice president and chief scientist of Natural Selection, Inc., in La Jolla, CA. The firm uses computational intelligence techniques to solve difficult real-world problems in industry, medicine, and defense. Dr. Fogel is the author of Evolutionary Computation: Toward a New Philosophy of Machine Intelligence, published by IEEE Press, 1995, as well as more than 90 publications in journals and conferences. Dr. Fogel is the first editor-in-chief of the IEEE Transactions on Evolutionary Computation, forthcoming May 1997, and is an associate editor for the IEEE Transactions on Neural Networks and the journal BioSystems, and is a member of the editorial boards of Evolutionary Computation and Fuzzy Sets & Systems.

Wednesday, November 13, 9:00 - 10:00 a.m.

Why Function Approximation Works So Well in Neural-Net Computing and How to Make it Work Even Better: A Critique and a Perspective

Yo-Han Pao, Case Western Reserve University, Cleveland, OH

Dr. Yoh-Han Pao is the Emeritus George S. Dively Professor of Engineering at Case Western Reserve University with appointments in Electrical Engineering and Computer Science, and President of AI Ware Inc., a software systems company, based on Computational Intelligence technologies. In his career at the University, Dr. Pao has served as Chairman of Electrical Engineering and as founder and Director of the Center for Automation and Intelligent Systems Research. Prior to his joining the University faculty, Dr. Pao had served for fourteen years in industry at Dupont and at the Bell Laboratories. He also served as Division Director of Electrical, Systems and Computer Engineering at the National Science Foundation during the years 1978-1980. Dr. Pao is a Fellow of IEEE and of the Optical Society of America. He has authored or co-authored many technical journal articles, chapters and books, including the 1989 Addison-Wesley book on Adaptive Pattern Recognition and Neural networks. He is the Founding Editor of the Quantum Electronics series of Academic Press. He serves on the Editorial board of several major technical journals. Dr. Pao has lectured and conducted research at various Universities and Institutes including MIT, Edinburgh University, Battelle Geneva and the Turing Institute. He earned his London University matriculation in 1939, and the Ph.D. degree from Pennsylvania State University in 1952.

BANQUET SPEAKER

Tuesday, November 12, 7:00 p.m.

Intelligent Control in Practice

Rahmat Shoureshi, Colorado School of Mines, Golden, CO

Professor Shoureshi completed his graduate studies at MIT in 1981. From 1981 to 1983, he was on the faculty of Wayne State University. In 1983, he joined the School of Mechanical Engineering at Purdue University where he was chairman of the Manufacturing and Materials Processing area (1988-1992), chairman of the Systems, Measurements and Control area (1992-1994), founder and director of the Center for Advanced Control of Energy and Power Systems (ACEPS). In August of 1994, he joined the Colorado School of Mines as the G.A. Dobelman Distinguished Professor of Engineering and Director of CSM-ACEPS.

Dr. Shoureshi has had an extensive involvement with teaching, new course/laboratory development, and initiating and leading new curricula. While at Wayne State University, he developed a new course and laboratory on microprocessors for measurement and control, and develop an advanced mechatronics course and laboratory at Purdue. His courses and laboratory experiments have been duplicated in several leading institutions nationally and internationally. At Purdue, he led the team in developing and incorporating manufacturing and materials processing curriculum, which is presently a very strong program at the School of Mechanical Engineering of Purdue University. Since joining CSM in 1994, Dr. Shoureshi has chaired the newly-established graduate program in Engineering Systems with three specialty areas of Energy and Power Systems with three specialty areas of Energy and Power Systems, Advanced Sensing and Automation, and Earth Systems. He has been the key organizer for the five sessions on education of Engineering Systems to be held at the 1997 ASME International Mechanical Engineering Conference and Exhibition ('97-IMECE). Dr. Shoureshi has given several short courses nationally and internationally on subjects and issues related to course and/or curriculum developments.

Dr. Shoureshi has had extensive research activities in the areas of intelligent control systems, active noise and vibration control, automation of manufacturing processes, mechatronics and system design, and control of energy and power systems. Results of over forty graduate students' theses have appeared in over 150 technical publications. In 1987, he received the American Automatic Control Council's Eckman Award for his outstanding contributions to the field. Dr. Shoureshi's research has been involved with close collaborations with several industries. He has served on several NSF and EPRI panels and has been on the advising boards of several industries. He is nationally and internationally known for his work in the area of intelligent control systems and active vibration and noise control. Dr. Shoureshi serves on editorial boards of different journals, he is the 1996-1997 chairman of the ASME Dynamic Systems & Control Division, and vice-chairman of the International Federation of Intelligent Automation.

LUNCHEON SPEAKERS

Monday, Noon - 1:30 p.m.

A Unified Framework for Neural Networks, Genetic Algorithms, and Their Engineering Applications

Erol Gelenbe, Duke University, Durham, NC

Erol Gelenbe's interests span neural networks and image processing, discrete stochastic processes, computer performance evaluation and computer networks, and genetic algorithms. He is currently active in publishing journal and conference papers in all of these areas. He is the Nello L. Teer Jr. Professor of Electrical and Computer Engineering and Chairman of his Department at Duke University. He was elected a Fellow of the IEEE at the age of 41 in 1986 for "leadership in the development of computer performance evaluation." His honors include the recent Grand Prix France Telecom (1996) of the French Academy of Science, the Science Award of the Parlar Foundation (Turkey, 1995), the Chevalier de l'Ordre du Merite (1992) from the French Government, and the IFIP Silver Core Award (1980). He is a graduate of the Middle East Technical University (Ankara, Turkey) in Electrical Engineering, and holds the Ph.D. (EE) from the Polytechnic Institute of Brooklyn and the D.Sc. (Applied Mathematics) from the University of Paris. Recently his research has been funded by IBM, NSF, ARO and the U.S. Army, NATO and the European Community's ESPRIT Program.

Tuesday, Noon - 1:30 p.m.

Fractal Behavior of the Electrocardiogram: Distinguishing Heart-Failure and Normal Patients using Wavelet Analysis

Malvin C. Teich, Boston University, Boston, MA

Dr. Malvin C. Teich has been a member of the faculty at Boston University since 1995. He holds appointments in the Departments of Electrical and Computer Engineering, Physics, and Biomedical Engineering. He conducts research in the areas of fractal point processes in physical and biological systems, information transmission in sensory systems, and photonics. His first professional association, in 1965, was with MIT Lincoln Laboratory, where he was engaged in work on infrared heterodyne detection. In 1967 he joined the faculty at Columbia University, where he served as a member of the Electrical Engineering Department (as Chairman from 1978 to 1980), the Applied Physics Department, and the Columbia Radiation Laboratory. Dr. Teich received the B.S. degree in physics from the Massachusetts Institute of Technology, the M.S. degree in electrical engineering from Stanford University, and the Ph.D. degree in electrical engineering from Cornell University. He is a Fellow of the Institute of Electrical and Electronics Engineers, the Optical Society of America, the American Physical Society, the American Association for the Advancement of Science, and the Acoustical Society of America. He is a member of Sigma Xi, Tau Beta Pi, the Biomedical Engineering Society, and the Association for Research in Otolaryngology. In 1969 he received the IEEE Broder J. Thompson Memorial Prize. Dr. Teich was awarded a Guggenheim fellowship in 1973. In 1992 he was honored with the Memorial Gold Medal of Palacky University in the Czech Republic. He is a Member of the Scientific board of the Czech Academy of Sciences Institute of Physics. He has authored or coauthored some 200 technical publications and holds one patent. He is the coauthor of Fundamentals of Photonics (Wiley, 1991).

Wednesday, Noon - 1:30 p.m.

Fusion of Fuzzy and Chaos, and its Application in Industrial Field: Time Series Prediction

Tadashi Iokibe, Meidensha Corporation, Tokyo, JAPAN

Dr. T. Iokibe is a Senior Engineer at Meidensha Corporation. Dr. Iokibe manages the research program in industrial applications applied fuzzy theory, GMDH, genetic algorithm and chaos theory. Dr. Iokibe received his B.E. degree from Osaka Institute of Technology, and Ph. D. from Hiroshima University. He has published or presented about 80 technical papers, and has been a director of SOFT (Japan Society for Fuzzy Theory and Systems) from May 1993 to July 1995, and is a chairman of the Kanto section of SOFT after April 1995. Dr. Iokibe has been awarded the Ohm award in 1991. He also is a member of IEEE, IEEJ, SICE and SOFT.

PLENARY SESSION I

MONDAY, NOVEMBER 11, 1996

8:30 - 8:40 a.m. Opening Remarks

Cihan H. Dagli, Conference Chairman

8:40 - 9:00 a.m. Welcoming Address

9:00 - 10:00 a.m. Plenary Address

Neural Networks for Control: Theory and Practice

Kumpati S. Narendra, Yale University, New Haven, CT

Monday, 10:30 a.m. - Noon

MA2.1 LEARNING ALGORITHMS AND TRAINING I

SESSION CHAIR: Yukinori Kakazu, Hokkaido University, Sapporo, JAPAN

Flexible Modular Architecture for Changing Environments

Viswanath Ramamurti and Joydeep Ghosh, The University of Texas at Austin, Austin, TX

Vibrating Potential Field for Large Scale Traveling Salesman Problem

Hiroshi Yokoi, Hokkaido University, Sapporo, JAPAN; Takafumi Mizuno and Masatoshi Takita, National Institute of Bioscience and Human Technology, JAPAN; and Yukinori Kakazu, Hokkaido University, Sapporo, JAPAN

Stability Analysis of Parametric Uncertain Neural Networks

Anke Meyer-Bäse, Technical University Darmstadt, Darmstadt, GERMANY

The Introduction of Principles of Elementary Quantum Theory in a Hopfield-style Network and the Solution of Constraint Satisfaction and Variable Binding Problems

Philip R. Van Loocke, University of Ghent, BELGIUM

A Simple Heuristic for Dynamic Network Design

Antony Hammitt and Eric B. Bartlett, Iowa State University, Ames, IA

MA2.2 FUZZY SYSTEMS I (FUZZY LOGIC)

SESSION CHAIR: Yasuhiko Dote, Muroran Institute of Technology, Muroran, JAPAN

General Parameter Neural Networks with Fuzzy Self-Organization

Daouren Akhemetov and Yasuhiko Dote, Muroran Institute of Technology, JAPAN

FUZZ: A Fuzzy-Based Concept Formation System that Integrates Human Categorization and Numerical Clustering

C. L. Philip Chen and Yuan Lu, Wright State University, Dayton, OH

Design and Application of Fuzzy-Logic-Based Gray Prediction Controller

Chin-Ming Hong and Chih-Tang Chiang, National Taiwan Normal University; and Ching-Tsan Chiang, Chien Hsin College of Technology and Commerce, Taiwan, R.O.C.

Application of Fuzzy Neural Hybrid Networks to State Prediction

Faridoon Shabani, Nadipuram R. Prasad, and Howard A. Smolleck, New Mexico State University, Las Cruces, NM

VLSI Implementation of a Universal Fuzzy Approximator

Bogdan M. Wilamowski, University of Wyoming, Laramie, WY; and Richard C. Jaeger, Auburn University, Auburn. AL

Monday, 10:30 a.m. - Noon

MA2.3 EVOLUTIONARY PROGRAMMING I

SESSION CHAIR: Ralf Salomon, University of Zürich, Zürich, SWITZERLAND

Optimal Sampling for Genetic Algorithms

Brad L. Miller and David E. Goldberg, University of Illinois, Urbana/Champaign, IL

Gesture-Based Programming, Part 1: The Approach

Richard M. Voyles and Pradeep K. Khosla, Carnegie Mellon University, Pittsburgh, PA

Gesture-Based Programming, Part 2: Primordial Learning

Richard M. Voyles, J. Daniel Morrow, and Pradeep K. Khosla, Carnegie Mellon University, Pittsburgh, PA

Evolving and Optimizing Braitenberg Vehicles

Ralf Salomon, University of Zürich, Zürich, SWITZERLAND

Constrained Genetic Programming with lil-gp

Cezary Z. Janikow, University of Missouri-St. Louis, St. Louis, MO

MA2.4 SMART ENGINEERING SYSTEMS I (ROBOTICS)

SESSION CHAIR: Nicholas G. Odrey, Lehigh University, Bethlehem, PA

A Study on Autonomous Robot with Empirical Measure

Jun Hakura, Hiroshi Yokoi, and Yukinori Kakazu, Hokkaido University, Sapporo, JAPAN

On the Application of Neural Networks to a Petri Net-Based Intelligent Workstation Controller for Manufacturing

Yi-Hui Ma, and Nicholas G. Odrey, Lehigh University, Bethlehem, PA

Neuro-Fuzzy Control of a Robotic Arm

Wallace Kelly III, Rajab Challoo, Robert A. McLauchlan, and S. Iqbal Omar, Texas A&M University-Kingsville, Kingsville, TX

A Multi-Neural Network Intelligent Path Planner for a Robot Arm

James P. Johnson, Rajab Challoo, Robert A. McLauchlan, and S. Iqbal Omar, Texas A&M University-Kingsville, Kingsville, TX

CMAC Hardware Design with Verilog Simulation

C.Y. Wu and C.S. Lin, University of Missouri-Columbia, Columbia, MO

Monday, 1:30 - 3:00 p.m.

MP1.1 LEARNING ALGORITHMS AND TRAINING II

SESSION CHAIR: Jari Vaario, Nara Women's University, Nara, JAPAN

A Binary Three Layered Neural Network with a Switched Error Perturbation learning and Its Iterative Learning Utilizing Generalization Property

Yohtaro Yatsuzuka, and Masaru Enomoto, (KDD) R & D Labs, JAPAN

Modular Learning in Multiagent Environment

J. Vaario, Nara Women's University, Nara, JAPAN; and Kanji Ueda, Kobe University, Kobe, JAPAN

A General Auto-Associative Memory with Sample Learning Capability

Hongchi Shi, University of Missouri-Columbia, Columbia, MO; Yunxin Zhao, University of Illinois, Urbana-Champaign, IL; and Xinhua Zhuang, University of Missouri-Columbia, Columbia, MO

Rapid and Reliable Learning through Data-Driven Initialization

Ralf Salomon, University of Zürich, Zürich, SWITZERLAND

Rule Set Quality Measures for Inductive Learning Algorithms

Ralf H. Klinkenberg, University of Missouri-Rolla, Rolla, MO; Daniel C. St. Clair, University of Missouri-Rolla, St. Louis, MO

Monday, 1:30 - 3:00 p.m.

MP1.2 FUZZY SYSTEMS II (CONTROL)

SESSION CHAIR: Roland Priemer, University of Illinois at Chicago, Chicago, IL

Design of a Fuzzy Controller Based on Fuzzy Closed-Loop Specifications

Cezary Kolodziej, and Roland Priemer, University of Illinois at Chicago, Chicago, IL

Fuzzy Logic Control of a Four-Link Robotic Manipulator in a Vertical Plane

Yuzhou Zhou, Robert A. McLauchlan, Rajab Challoo, and S. Iqbal Omar, Texas A & M University, Kingsville, TX

Application of the Fuzzy State Fuzzy Output Finite State Machine to the Problem of Recovery from Violation of Ontological Assumptions

Janos L. Grantner, Western Michigan University, Kalamazoo, MI; George Fodor, ABB Industrial Systems AB, SWEDEN; Dimiter Driankov, Linkoping University, SWEDEN; and Marek J. Patyra, University of Minnesota, Duluth, MN

Inverse-Model Compensation Using Fuzzy Modeling and Fuzzy Learning Schemes

P.J. Costa Branco and J.A. Dente, Instituto Superior Técnico, PORTUGAL

Fuzzy Control Design by Phase Portrait Analysis: An Application to Exothermic Chemical Reactor Control

Zhongjun Yu and Degang Chen, Iowa State University, Ames, IA

MP1.3 EVOLUTIONARY PROGRAMMING II

SESSION CHAIR: Anna L. Buczak, AlliedSignal, Inc., Morristown, NJ

Solving Large Scale Block Stacking Problem: A Distributed Approach

Masaaki Minagawa, Sapporo Gakuin University, JAPAN; and Yukinori Kakazu, Hokkaido University, Sapporo, JAPAN

Genetic Algorithm-Based Method for Solving an Over determined Set of Equations Corrupted by Measurement Noise

Anna L. Buczak, and Joseph J. Barrett, AlliedSignal, Inc., Morristown, NJ

Optimizing the Growth of PECVD Silicon Nitride Films for Solar Cell Applications Using Neural Networks

Seung-Soo Han, Li Cai, Ajeet Rohatgi, and Gary S. May, Georgia Institute of Technology, Atlanta, GA

Hydraulic System Parameter Identification Using Genetic Algorithm Methods

Ruth Book, University of Illinois at Urbana-Champaign, Urbana, IL

Testing Reusable Software Components by Combining Genetic Algorithms and Backpropagation Algorithms

Junda Chen and David C. Rine, George Mason University, Fairfax, VA

MP1.4 SMART ENGINEERING SYSTEMS II ( PROCESS MONITORING)

SESSION CHAIR: John P.H. Steele, Colorado School of Mines, Golden, CO

Detecting Cavitation in Hydraulic Pumps Using Artificial Neural Networks

John P.H. Steele, Michelle Archuleta, and Tyler Hooley, Colorado School of Mines, Golden, CO

Classifying the Quality of Wire Bonds: Neural Networks Versus Discriminant Analysis

Xiaoyun Sun, BehavHeuristics, Inc., Hanover, MD; Bruce Golden, University of Maryland, College Park, MD; and Edward Wasil, American University, Washington, D.C.

Evaluation of Model Discrimination Techniques in Artificial Neural Networks with Application to Grain Drying

Angelita D. N. Garth, Iowa State University, Ames, IA; Victoria C.P. Chen, Georgia Institute of Technology, Atlanta, GA; Jun Zhu and Derrick K. Rollins, Iowa State University, Ames, IA

Anti-Collision Braking System

Naveen Ramagiri and Peng-Yung Woo, Northern Illinois University, DeKalb, IL

Using SOFM Neural Networks for Part Identification in Automated Inspection System

Kuang-Han Hsieh, and C. Alec Chang, University of Missouri-Columbia, Columbia, MO

Monday, 3:30 - 5:00 p.m.

MP2.1 LEARNING ALGORITHMS AND TRAINING III

SESSION CHAIR: Bogdan M. Wilamowski, University of Wyoming, Laramie, WY

Implementation of RBF Type Networks by MLP Networks

Bogdan M. Wilamowski, University of Wyoming, Laramie, WY

Noisy Recurrent Neural Networks

Soumitra Das and Oluseyi Olurotimi, George Mason University, Fairfax, VA

Estimating the Solvability of Pattern Classification Problems

S. Haring, Utrecht University, NETHERLANDS; J.N. Kok, Leidden University, NETHERLANDS; M.A. Viergever, Utrecht University Hospital, NETHERLANDS

The Analysis and Application for Gray Correlation Theory

Chin-Ming Hong, and Terng-Chiao Lin, National Taiwan Normal University, Taiwan, R.O.C.; Ching-Tsan Chiang, Chien Hsin College of Technology and Commerce, Taiwan, R.O.C.; Lon-Biou Lin, National Taiwan Normal University, Taiwan, R.O.C.

A Novel Framework for Neural Net Learning

Tsu-Shuan Chang, University of California, Davis, CA; and Jenq-Neng Hwang, University of Washington, Seattle, WA

MP2.2 FUZZY SYSTEMS III (CONTROL)

SESSION CHAIR: Ernest L. Hall, University of Cincinnati, Cincinnati, OH

Fuzzy Logic Processing and Dynamic Alarm Handling for Real-Time Machine Health Monitoring

Tim D. Seifert, and John P.H. Steele, Colorado School of Mines, Golden, CO

Model Reference Adaptive Fuzzy Controller Design Using Genetic Algorithms

Wen-Ruey Hwang, Saleh Zein-Sabatto, and John G. Kuschewski, Tennessee State University, Nashville, TN

Fuzzy Logic System for Three Dimensional Line Following for a Mobile Robot

Tayib Samu, Nikhil Kelkar, and Ernest L. Hall, University of Cincinnati, Cincinnati, OH

Robot Control Using Fuzzy Logic and Neural Network

Beheshteh Sayeh, Southern Illinois University at Carbondale, Carbondale, IL

Fuzzy Logic Control of FMS Conveyor System

Eman Kamel and William Biles, University of Louisville, Louisville, KY

MP2.3 EVOLUTIONARY PROGRAMMING III

SESSION CHAIR: K. C. Chan, University of New South Wales, Sydney, AUSTRALIA

VLSI Placement Using Genetic Algorithms

K. W. Leavitt, A. R. Marudarajan, and . N. Nouhi, and H. N. Riley, California State Polytechnic University, Pomona, CA

Design of 2-Dimensional Crossover Operators for the Facilities Layout Problem

Sulung and K. C. Chan, University of New South Wales, Sydney, AUSTRALIA

Spatial Layout Planning Using Evolved Design Genes

John S. Gero and Vladimir A. Kazakov, The University of Sydney, Sydney, AUSTRALIA

Assembly Line Balancing Using a Genetic Algorithm with Repair Routine

K. C. Chan, and R. Tantono, University of New South Wales, Sydney, AUSTRALIA

Job Shop Scheduling Problem Revisited: An Efficient Unification of Genetic Algorithm with Simulated Annealing

Sabyasachi Ghoshray, Florida International University, Miami, FL

Monday, 3:30 - 5:00 p.m.

MP2.4 SMART ENGINEERING SYSTEMS III (PROCESS MONITORING)

SESSION CHAIR: Ratna Babu Chinnam, North Dakota State University, Fargo, ND

Performance Reliability Prediction: Estimating Confidence Intervals for FFNs Using Self-Organizing Feature Maps

Ratna Babu Chinnam and Nikhil Shrikhande, North Dakota State University, Fargo, ND

Rapid Identification of Nuclear Power Plant Malfunctions with Artificial Neural Networks via Fourier Transformed Signals

Sandor Benedek, and Mark J. Embrechts, Rensselaer Polytechnic Institute, Troy, NY

Artificial Neural Network and the Taguchi Method Application for Optimum Ultrasonic Welding Process Design

Jung Eui Hong, Mando Institute of Production Technology; Gerald W. Greenway, and Kenneth M. Ragsdell, University of Missouri-Rolla, Rolla, MO

Performance Monitoring System of Valves Operation

M. H. Wu, University of Derby, UNITED KINGDOM

The Application of Neural Networks to a Condition Monitoring System for the Optical Fibre Drawing Process

H. Shi, C. Tite, A. D. Hope, S. Noroozi, Southampton Institute, Southampton, UNITED KINGDOM

PLENARY SESSION II

TUESDAY, NOVEMBER 12, 1996

9:00 - 10:00 a.m. Plenary Session

Evolutionary Computation: The New Synthesis

David Fogel, Natural Selection, Inc., La Jolla, CA

Tuesday, 10:30 a.m. - Noon

TA2.1 LEARNING ALGORITHMS AND TRAINING IV

SESSION CHAIR: R. J. Mears, Cambridge University, Cambridge, U.K.

Approaches for Symbolically Interpreting Artificial Neural Networks

Ismail Taha and Joydeep Ghosh, University of Texas, Austin, TX

A Study on State Grouping and Opportunity Evaluation for Reinforcement Learning Methods

Wenwei Yu, Hiroshi Yokoi and Yukinori Kakazu, Hokkaido University, Sapporo, JAPAN

Neural Networks in the Context of Autonomous Agents: Important Concepts Revisited

Ralf Salomon, University of Zürich, Zürich, SWITZERLAND

Backpropagation Learning Using Positive Weights in Optoelectronic Neural Networks

W.M.D. Bradley, and R.J. Mears, Cambridge University, Cambridge, UNITED KINGDOM

Predicting Local Minima for Weight Training of Backpropagation Nets

Wei Cao, and James H. Burghart, Cleveland State University, Cleveland, OH

Tuesday, 10:30 a.m. - Noon

TA2.2 FUZZY SYSTEMS IV (PATTERN RECOGNITION)

SESSION CHAIR: Hartmut Ewald, Hochschule, Wismar, Wismar, GERMANY

A Fuzzy Logic Classifier for the Detection of Bounded Weak Echo Regions in Meteorological Images

V. Lakshmanan, University of Oklahoma, OK; Arthur Witt, National Severe Storms Laboratory, Norman, OK

A Fuzzy Technique for Flow Oriented Image Direction Computing

D.C. Douglas Hung and Ching-Yu Huang, New Jersey Institute of Technology, Newark, NJ

Classification of Eddy Current Signals Using Fuzzy-Logic

Hartmut Ewald, and Michael Stieper, Hochschule Wismar, GERMANY

Comparison of the Performance of the Neural Network- and Fuzzy Logic-Based Decision Units for a Robot Safety System

Jozef Zurada and Andrew L. Wright, University of Louisville, KY

An ANN Architecture for the Fusion of Fuzzy Pattern Recognition Systems

Esin Ulug, Intelligent Neurons Inc., Deerfield Beach, FL

TA2.3 EVOLUTIONARY PROGRAMMING IV

SESSION CHAIR: J.N. Carter, Imperial College, London, U.K.

Optimum Performance of Genetic Algorithms and Derivatives of Simulated Annealing for Combinatorial Optimisation

J.N. Carter, Imperial College, London, UNITED KINGDOM

Using Genetic Algorithms to Design Combinational Logic Circuits

Carlos A. Coello Coello, Alan D. Christiansen, Arturo Hernández Aguirre, Tulane University, New Orleans, LA

Application of a Modified Genetic Algorithm to Parameter Estimation in the Petroleum Industry

M. Bush, Amoco services Ltd, London, UNITED KINGDOM; J.N. Carter, Imperial College, London, UNITED KINGDOM

Application and Evaluation of Genetic Programming for Aimpoint Selection

Carey Schwartz, Charles Keyes, and Erik van Bronkhorst, Naval Air Warfare Center Weapons Division, China Lake, CA

Directed Mutation Applied to Nonlinear Programming

Francisco Antonio Pereira Fialho, Universidade Federal de Santa Caterina, Santa Caterina, BRAZIL

TA2.4 SMART ENGINEERING SYSTEMS IV (NOVEL APPLICATIONS)

SESSION CHAIR: Keith A. Osman, University of Central England in Birmingham, Birmingham, UNITED KINGDOM

A Hybrid Neural Network Based Pattern-Recognition Engine for Out-door Electronic Nose Application

B. Yang, and M. C. Carotta, University of Ferrara, Ferrara, ITALY; G. Faglia, University of Brescia, Brescia, ITALY; M. Ferroni, and V. Guidi, and G. Martinelli, and G. Sberveglieri, University of Ferrara, Ferrara, ITALY

Quantification of Shape Characteristics by Using Neural Nets

Sanjay Goel, General Electric Corporate Research and Development, Schenectady, NY; Mark Embrechts, Rensselaer Polytechnic Institute, Troy, NY

Optimised Control of Multiple Power Sources Using Genetic Algorithms, Neural Networks and Fuzzy Logic: A Feasibility Study

K. A. Osman, and A. M. Higginson, University of Central England in Birmingham, Birmingham, UNITED KINGDOM

Neural Networks for Accelerating the Transistor Design Process

Ryan Ferguson, and David J. Roulston, University of Waterloo, Waterloo, CANADA

Separation of Hashish Signatures with Neural Networks

A. Tenhagen, Th. Feuring, W.-M. Lippe, Universität Münster, Münster, GERMANY; H. Lahl, G. Henke, Universität Münster, Münster, GERMANY

Tuesday, 1:30 - 3:00 p.m.

TP1.1 LEARNING ALGORITHMS AND TRAINING V

SESSION CHAIR: K. KrishnaKumar, The University of Alabama, Tuscaloosa, AL

Training of Artificial Neural Networks via a Barrier Function Method with L1 Norm

Theodore B. Trafalis and Tarek A. Tutunji, University of Oklahoma, Norman, OK

Adaptive Optimal Filtering by Neural Networks with Long- and Short-Term Memories

James Tin-Ho Lo, University of Maryland, Baltimore, MD

Non-Synaptic Learning in CA1 Neurons

Wayne C. Jouse, University of Arizona, Tucson, AZ

Modeling the Amacrine Cells in the Primate Retina for Improved Edge Detection and Noise Reduction of Images Provided to Artificial Vision Systems

David Enke, and Cihan H. Dagli, University of Missouri-Rolla, Rolla, MO

Immunized Artificial Neural Systems as Adaptive Critics

K. KrishnaKumar, J. Neidhoefer, R. Kumaralingam, and K. Nishita, The University of Alabama, Tuscaloosa, AL

TP1.2 PATTERN RECOGNITION I

SESSION CHAIR: Mark J. Embrechts, Rensselaer Polytechnic Institute, Troy, NY

Buffered Reset Leads to Improved Compression in Fuzzy ARTMAP Classification of Radar Range Profiles

Stephen Grossberg, Mark A. Rubin, and William W. Streilein, Boston University, Boston, MA

Spatial Spectral Image Analysis Using ART Networks

Sophia Roberts, Florida State University, Tallahassee, FL; Galen R. Gisler, and James Theiler, Los Alamos National Laboratory, Los Alamos, NM

IMAGINET: A Novel Probabilistic Neural Network for Rapid Multiscale Image Classification

Mark J. Embrechts, Russell P. Kraft, Vijay Sankaran, and Don L. Millard, Rensselaer Polytechnic Institute, Troy, NY

Comparison of Three Clustering Algorithms and an Application to Color Image Compression

Jihun Cha, and Laurene V. Fausett, Florida Institute of Technology, Melbourne, FL

Fuzzy Binarization and Segmentation of Text Images

Mario I. Chacon M., New Mexico State University, Las Cruces, NM

TP1.3 BIO-MEDICAL ENGINEERING I

SESSION CHAIR: Daniel Graupe, University of Illinois, Chicago, IL

Neural Net Classification of REM Sleep Based on Spectral Measures as Compared to Nonlinear Measures

Jürgen Fell, Joachim Röschke, Michael Grözinger, University of Mainz, Mainz, GERMANY

Therapeutic REM Sleep deprivation Rendered Possible by Application of Artificial Neural Networks to One Channel EEG Data

M. Grözinger, T. Uhl, C. Schäffner, and J. Röschke, University of Mainz, Mainz, GERMANY

Mental Workload Classification Using a Backpropagation Neural Network

Chris A. Russell, Corrina T. Monett, and Glenn F. Wilson, A1/CFHP, Wright-Patterson AFB, OH

A Preliminary Investigation of Selection of EEG and Psychophysiological Features for Classifying Pilot Workload

Greene, Bauer, Kabrisky, Rogers, Russell, and Wilson, Air Force Institute of Technology, Wright-Patterson AFB, OH

Classification of Movement Patterns in Patients with Low Back Pain Using an Artificial Neural Network

J. B. Bishop, and M. H. Pope, University of Iowa, Iowa City, IA; M. Szpalski, Brussels Free University, Brussels, BELGIUM; and

S. Ananthraman, Neural Applications Corporation, Coralville, IA

Neural Network for Control of Stimulation of Peripheral Motor Neurons to Allow Patient Responsive Ambulation by Paraplegics

Daniel Graupe, and Hubert Kordylewski, University of Illinois, Chicago, IL

Tuesday 1:30 - 3:30 p.m.

TP1.4 SMART ENGINEERING SYSTEMS V (MANUFACTURING PROCESSES)

SESSION CHAIR: Janet M. Twomey, Wichita State University, Wichita, KS

A Neural Network Predictor for Reengineering of Resin Manufacturing

Yu-To Chen, Pratap S. Khedkar, GE Research and Development, Schenectady, NY; and Maarten P. Ter Weeme, GE Plastics,

Mt. Vernon, IN

Parametric Determination of W-M Function for Machined Surface Characterization

Guang Yang, Graham Smith, Caroline Tite, and Tony Hope, Southampton Institute, Southampton, UNITED KINGDOM

An Investigation of the Sensitivity of Perceptron Neural Networks to Tool Wear Inception During a Turning Process

Dimla E. Dimla, Jnr., Paul M. Lister, and Nigel J. Leighton, University of Wolverhampton, Wolverhampton, UNITED KINGDOM

Application of Neural Network in the Analysis of Rolling Processes

Zhengjie Jia, Jay S. Gunasekera, Luis C. Rabelo, and Mal Gunasekera, Ohio University, Athens, OH

Artificial Neural Network Approach to the Control of a Wave Soldering Process

Janet M. Twomey, Wichita State University, Wichita, KS; and Alice E. Smith, University of Pittsburgh, Pittsburgh, PA

Tuesday, 3:30 - 5:00 p.m.

TP2.1 LEARNING ALGORITHMS AND TRAINING VI

SESSION CHAIR: Alice E. Smith, University of Pittsburgh, Pittsburgh, PA

Ontogenesis in Neural Networks with Feedback

Thomas S. Dranger, and Roland Priemer, University of Illinois, Chicago, IL

Dynamical Behavior of Artificial Neural Networks with Random Weights

D. J. Albers, and J. C. Sprott, University of Wisconsin, Madison, WI; and W. D., Dechert, University of Houston, Houston, TX

Iterative Training to Improve Neural Network Metamodels of Simulated Systems

Saniye Burcu Ozserim and Alice E. Smith, University of Pittsburgh, Pittsburgh, PA; and Robert A. Kilmer, U.S. Military Academy, West Point, NY

Hierarchical Neural Network Architectures for Deployment in Multi-Decision Environments

Belur V. Dasarathy, Dynetics, Inc., Huntsville, AL

Error Phenomena of Backpropagation Learning

Yuadkoun Dhiantravan and Roland Priemer, University of Illinois, Chicago, IL

TP2.2 PATTERN RECOGNITION II

SESSION CHAIR: Ryan G. Rosandich, University of Minnesota-Duluth, Duluth, MN

Identification of the Well Test Interpretation Model Using the Extended HaVNet Neural Network

Edward A. May, and Cihan H. Dagli, University of Missouri-Rolla, Rolla, MO

Object Segmentation by Stereo Disparity: A Neural Approach

Ryan G. Rosandich, University of Kansas Regents Center, Overland Park, KS

A Neural Network Based Classification of Bongard Patterns

Lakshmi N. Srinivasa, Ernest L. Hall, and Sam Anand, University of Cincinnati, Cincinnati, OH

Determination of Carbonate Lithofacies Using Heirarchical Neural Networks

Haiming Yang, H. C. Chen, and J. H. Fang, The University of Alabama, Tuscaloosa, AL

GBF Network Architectures for Robot Vision

Josef Pauli, Christian-Albrechts-Universität, Kiel, GERMANY

Tuesday, 3:30 - 5:00 p.m.

TP2.3 BIO-MEDICAL ENGINEERING II

SESSION CHAIR: Colin W. Morris, University of Glamorgan, Pontypridd, UNITED KINGDOM

Classification as Unknown by RBF Networks: Discriminating Phytoplankton Taxa from Flow Cytometry Data

Colin W. Morris, University of Glamorgan, Pontypridd, UNITED KINGDOM; and Lynne Boddy, University of Wales, Cardiff, UNITED KINGDOM

The Growth of Escherichia coli 0157:H7: A Backpropagation Network Approach

Maha N. Hajmeer, Imad A. Basheer, and Yacoub M. Najjar, Kansas State University, Manhattan, KS

A Study on Image Analysis of Amoebae from Phase-Contrast Microscope Image

Masao Kubo, Hiroshi Yokoi, and Yukinori Kakazu, Hokkaido University, Sapporo, JAPAN

Wavelets for Shape Recognition with Applications to Mammography

L. M. Bruce and R. R. Adhami, University of Alabama in Huntsville, Huntsville, AL

A Large Memory Storage and Retrieval Neural Network for Browsing and Medical Diagnosis Application

Daniel Graupe, and Hubert Kordylewski, University of Illinois, Chicago, IL

TP2.4 SMART ENGINEERING SYSTEMS VI (GENERAL APPLICATIONS)

SESSION CHAIR: Yacoub M. Najjar, Kansas State University, Manhattan, KS

Coding a Conceptual Model into a Neural Network in Modeling of Ice-Correction

Markus Huttunen, Finnish Environment Institute, Helsinki, FINLAND; Esko Ukkonen, University of Helsinki, Helsinki, FINLAND; and Bertel Vehviläinen, Finnish Environment Institute, Helsinki, FINLAND

Neural Network Applications in Rainfall-Runoff Modeling

J. Anmala, R. S., Govindaraju, R. Mysore, and B. Zhang, Kansas State University, Manhattan, KS

A Neural Network-Based Distress Model for Kansas JPCP Longitudinal Joints

Imad A. Basheer, and Yacoub M. Najjar, Kansas State University, Manhattan, KS; and John Wojakowski, Kansas Department of Transportation, Topeka, KS

Deformation Prediction of Underground Excavations at WIPP Using Backpropagation

Sangki Kwon, and Hamish Miller, University of Missouri-Rolla, Rolla, MO

Genetic and Generalized Bar-Delta Neural Network for Assessing Transient Stability of a Power System

M. Moechtar, L. Hu, and T. C. Cheng, University of Southern California, Los Angeles, CA

Utilizing Neural Networks to Predict Wing Leading Edge Slats Airloads

Elias Bounajem, and Alan VanDerVeen, Cessna Aircraft Co., Wichita, KS

PLENARY SESSION III

WEDNESDAY, NOVEMBER 13, 1996

9:00 - 10:00 a.m. Plenary Address

Why Function Approximation Works So Well in Neural-Net Computing and How to Make it Work Even

Better: A Critique and a Perspective

Yo-Han Pao, Case Western Reserve University, Cleveland, OH

Wednesday, 10:30 - Noon

WA2.1 CONTROL I

SESSION CHAIR: Donald C. Wunsch, Texas Tech University, Lubbock, TX

Neural Network Controllers: A Literature Review

Benito Fernández-Rodríguez, Nelson Jaramillo-Mendez, and Nirad Pandya, University of Texas at Austin, Austin, TX

Neuro-Gain Approximation: A New Approach to Real Time Adaptive Neurocontrol (A Continuous Approximation to the Nonlinear Mapping Between Linear Controllers)

J. Neidhoefer, and K. KrishnaKumar, The University of Alabama, Tuscaloosa, AL

Neurocontrollers for Ball-and-Beam Systems

Paul H. Eaton, Danil V. Prokhorov, and Donald C. Wunsch, II, Texas Tech University, Lubbock, TX

Control of a Nonlinear Multivariable System with Adaptive Critic Designs

Nikita A. Visnevski, and Danil V. Prokhorov, Texas Tech University, Lubbock, TX

Real-Time Learning of Aircraft Parameters Using Recursive Least Squares to Train RBF Networks

J. Nicholas Laneman, Massachusetts Institute of Technology, Cambridge, MA; and Gerald E. Peterson, McDonnell Douglas Corporation, St. Louis, MO

WA2.2 PATTERN RECOGNITION III

SESSION CHAIR: Mansur R. Kabuka, University of Miami, Miami, FL

An Autonomous Boolean Neural Network Approach for Image Understanding

Essam A. El-Kwae and Mansur R. Kabuka, University of Miami, Miami, FL

Using Immunological Principles in Anomaly Detection

Dipankar Dasgupta, University of Missouri-St. Louis, St. Louis, MO

Classification of the Effects of F-actin under Treatment of Drugs in Endothelial Cells

Kamal J. Khiani, Sameh M. Yamany, and Aly A. Farag, William D. Ehringer and Fredrick N. Miller, University of Louisville,

Louisville, KY

3D Segmentation and Labeling Using Unsupervised Clustering for Volumetric Measurements on Brain CT Imaging to Quantify TBI Recovery

Mohamed N. Ahmed, and Aly A. Farag, University of Louisville, Louisville, KY

Neuro-Fuzzy Models for Mammography: Diagnosis of Breast Cancer

A. D. Kulkarni, N. R. Bakthula, and S. V. Boypati, University of Texas at Tyler, Tyler, TX

Wednesday, 10:30 - Noon

WA2.3 DATA ANALYSIS

SESSION CHAIR: Richard T. Gordon, Chubb Group of Insurance Companies, Liberty corner, NJ

A Draft Standard for the Certification of Neural Networks used in Safety Critical Systems

D. F. Bedford, J. Austin, and G. Morgan, University of York, York, UNITED KINGDOM

A New Chaos Based Approach to Data Conceptualization

Richard T. Gordon and Richard J. Cantor, Chubb Group of Insurance Companies, Liberty Corner, NJ; Mark J. Woyshville, Case Western Reserve University, Cleveland, OH

Data Analysis and Model Construction Using Constrained Categorical Regression

Harvey L. Bodine, Steven S. Henley, and Robert L. Dawes, Martingale Research Corporation, Allen, TX; Richard M. Golden, University of Texas at Dallas, Richardson, TX; and T. Michael Kashner, UT Southwestern Medical Center at Dallas, Dallas, TX

Discovering Relationships in Relational Databases Through Data Mining

David J. Harrier, Monsanto Agricultural Group; and Daniel C. St. Clair, University of Missouri-Rolla, St. Louis, MO

An Algorithm for Clustering Non-Deterministic Data

Charles Wittmaier, AAA Auto Club of Missouri, St. Louis, MO; and Chaman L. Sabharwal, University of Missouri-Rolla, St. Louis, MO

WA2.4 SMART ENGINEERING SYSTEMS VII (PREDICTION)

SESSION CHAIR: Alireza Khotanzad, Southern Methodist University, Dallas, TX

Machine Tool Failure Prediction by Chaos and Fuzzy Computing Approach

Takashi Nishimura, Dauren Akhmetov and Yasuhiko Dote, Muroran Institute of Technology, Muroran, JAPAN

Iterative Prediction of Chaotic time Series Using a Recurrent Neural Network

Magdi A. Essawy, and Mohammad Bodruzzaman, Tennessee State University, Nashville, TN; Abolghasem Shamsi, and Stephen Noel, Morgantown Energy Technology Center (METC), Morgantown, WV

A Confidence Measure for GRNN-Based System Modeling and Prediction

G. Dan Hammack, and Alireza Khotanzad, Southern Methodist University, Dallas, TX

Predict Building Construction Productivity Using Neural Networks

Jingsheng Shi, University of Hong Kong, Kowloon, HONG KONG

Prediction of Shear Stress-Strain Behavior of Soil with Recurrent Neural Network

Jian-Hua Zhu, Musharraf M. Zaman, and Theodore B. Trafalis, University of Oklahoma, Norman, OK

Wednesday, 1:30 - 3:00 p.m.

WP1.1 CONTROL II

SESSION CHAIR: K. Chandrashekhara, University of Missouri-Rolla, Rolla, MO

Robust Vibration Control of Composite Beams Using Piezoelectric Devices and Neural Networks

C. P. Smyser, S. Varadarajan, and K. Chandrashekhara, University of Missouri-Rolla, Rolla, MO

A New Control Scheme Applied to the Truck Backer Problem

Robert Woodley, Tianjing Han and Levent Acar, University of Missouri-Rolla, Rolla, MO

Backing Up a Truck and Trailer Using Sets of Three-Neuron Controllers

John R. Alexander, Jr., Towson state University, Towson, MD; and Jacob Thomas Cox, Vitro Corporation, Rockville, MD

Two Methods of Adaptive Controlled Channel Resource Allocation Using Reinforcement Learning and Supervised Learning Techniques

Edward J. Wilmes, Omnipoint Corporation, Colorado Springs, CO; and Kelvin T. Erickson, University of Missouri-Rolla, Rolla, MO

Backpropagation for Spike Sorting in Synthetic Extracellular Recordings

Ismael Espinosa, Jorge Quiza, Luis Rivera, Irma Domínguez, Mexico National University, Unam, MEXICO; and Ritaluz Lara, Mexico National Polytechnic Institute (IPN), MEXICO

Wednesday, 1:30 - 3:00 p.m.

WP1.2 PATTERN RECOGNITION IV

SESSION CHAIR: Sebastiaan Haring, Utrecht University, Utrecht, THE NETHERLANDS

Tuning Image Filters Using Feed-Forward Networks

S. Haring, Utrecht University, Utrecht, THE NETHERLANDS; J. N. Kok, Leiden University, Leiden, THE NETHERLANDS; and M. A. Viergever, Utrecht University Hospital, Utrecht, THE NETHERLANDS

Unsupervised Cluster Discovery Using Reconciliation

Larry K. Barrett and Peter Bock, The George Washington University, Washington, D.C.

Appropriate Scales When Using Wavelets for Feature Extraction

L. M. Bruce, R. R. Adhami, and J. W. Bruce, University of Huntsville, Huntsville, AL

A Fuzzy Contour Model

V. Vuwong, and L. S. Wilson, Ultrasonics Laboratory (CSIRO), Sydney, AUSTRALIA; J. Hiller, and J. Jin, The University of New South Wales, Sydney, AUSTRALIA

A Neural Network for Detecting Edges in an Image

Robert B. Heathcock, Top of the Gulf, Inc., Mexico Beach, FL

WP1.3 COMMUNICATIONS

SESSION CHAIR: Thomas L. Hemminger, Penn State University, Erie, PA

Adaptive Communication Channel Equalization by Neural Networks with Long- and Short-Term Memories

James Ting-Ho Lo, University of Maryland Baltimore County, Baltimore, MD

A Neural Network Solution to the Multicast Packet Radio Transmission Problem

Thomas L. Hemminger, Penn State University, Erie, PA; and Carlos A. Pomalaza-Raez, Indiana University-Purdue University,

Fort Wayne, IN

Optimising Information Transmission for Data Compression Using Feedforward Neural Networks

C. Whitrow, K. A. Osman, and A. M. Higginson, University of Central England, Birmingham, ENGLAND

An Improved Hopfield Network Approach to Channel Assignment in a Cellular Mobile Communications Network

Kate Smith, Monash University, Clayton, Victoria, AUSTRALIA; and M. Palaniswami, The University of Melbourne, Parkville, Victoria, AUSTRALIA

An Application of graph Theory to Communication Systems: A Case Study

Orhan Turkbey, Gazi University, Ankara, TURKEY

WP1.4 SMART ENGINEERING SYSTEMS VIII (PREDICTION)

SESSION CHAIR: Sabyasachi Ghoshray, Florida International University, Miami, FL

Periodic Time-Series Analysis Using Neural Networks

Senshu Ye, and C. L. Philip Chen, Wright State University, Dayton, OH

Hybrid Prediction Technique by Fuzzy Inferencing on the Chaotic Nature of Time Series Data

Sabyasachi Ghoshray, Florida International University, Miami, FL

Time Series Prediction Using Associative Memory Neural Networks

S. J. Silver-Warner, and R. J. Glover, Brunel University, Uxbridge, Middlesex, ENGLAND

Evolving Recurrent Bilinear Perceptrons for Time Series Prediction

Sathyanarayan S. Rao, and Kumar Chellapilla, Villanova University, Villanova, PA

Hybrid Box-Jenkins and Neural Network Forecasting of Potable Water Demand

Qin Wang and Miriam Heller, University of Houston, Houston, TX

Wednesday, 3:30 - 5:00 p.m.

WP2.1 CONTROL III

SESSION CHAIR: Levent Acar, University of Missouri-Rolla, Rolla, MO

A New Abstract Artificial Neural Architecture for Solving Difficult Learning Control Problems

Edward A. Thompson, and Georges A. Bécus, University of Cincinnati, Cincinnati, OH

Characterization of Contact Motion Kinematics with Recurrent Neural Networks

Chidambar Ganesh, David J. Ferkinhoff and Kai F. Gong, Naval Undersea Warfare Center, Newport, RI

Galerkin Expansion and Artificial Neural Network Based Low-Order Dynamical Modeling of a Channel Flow

R. A. Sahan, D. Albin Jr., and N. Sahan, Lehigh University, Bethlehem, PA

A Real-Time Neural Network Controller for Nonlinear Pneumatic Valve

J. F. Glenn, L. M. Schaina, S. Cheng, A. R. Marudarajan, Y. Cheng, M. S. El-Sawah and E. T. Ibrahim, University, Pomona, CA

A Method for Modeling the Parameters of the Non-Stationary Systems Using Neural Networks

Mihaela Cistelecan, Technical University of Cluj-Napoca, CLUJ-Napoca, ROMANIA

WP2.2 OPTIMIZATION

SESSION CHAIR: Theodore B. Trafalis, University of Oklahoma, Norman, OK

An Affine Scaling Scatter Search Approach for Continuous Global Optimization Problems

Theodore B. Trafalis and Suat Kasap, University of Oklahoma, Norman, OK

Using Artificial Neural Networks to Solve Generalized Orienteering Problems

Qiwen Wang, Peking University, Beijing, CHINA; Xiaoyun Sun, BehavHeuristics, Inc., Hanover, MD; and Bruce L. Golden, University of Maryland, College Park, MD

Intelligent Computing Budget Allocation for Stochastic Simulation

Chun-Hung Chen, University of Pennsylvania, Philadelphia, PA; and Liyi Dai, Washington University, St. Louis, MO

Financial Planning with Fuzzy Linear Programming

Zülal Güngör, Gazi University, Ankara, TURKEY

The Theory of Discrete Optimization and its Relationship with Facilities Layout

Orhan Turkbey, Gazi University, Ankara, TURKEY

WP2.3 DECISION ANALYSIS

SESSION CHAIR: Daniel C. St. Clair, University of Missouri-Rolla, EEC, St. Louis, MO

A Systematic Approach for Designing Neural Networks with Application to Classification, Decision, and Control

Nader M. Boustany, General Motors Energy Center, Troy, MI

Fuzzy Logic Approach for Serviceability Evaluation

Jiangping Wang, and Venkat Allada, University of Missouri-Rolla, Rolla, MO

Injection Mold Complexity Evaluation Model Using a Backpropagation Network Implemented on a Parallel Computer

Rawin Raviwongse and Venkat Allada, University of Missouri-Rolla, Rolla, MO

Sensitivity Analysis Validation for Expert Networks

Richard T. Gordon, Michael C. Jasman and Richard J. Cantor, Chubb Group of Insurance Companies, Liberty Corner, NJ

Using Grouping and Uncertain Reasoning during ID3 Decision Tree Construction and Testing

Carmela R. C. Santos, Electronic Data Systems Corporation, Maryland Heights, MO; Daniel C. St. Clair, University of Missouri-Rolla,

St. Louis, MO; and Peter E. Maher, Open Computing Institute, Inc., St. Louis, MO

A Table-Based Approach to Expert Fault Diagnosis Networks

Alan P. Levis, Kristin L. Adair, and Susan I. Hruska, Florida State University, Tallahassee, FL

Wednesday, 3:30 - 5:00 p.m.

WP2.4 SMART ENGINEERING SYSTEMS IX (PREDICTION)

SESSION CHAIR: Saleh M. Al-Alawi, Sultan Qaboos University, Sultanate of Oman, OMAN

A Seasonal and Modular Network Approach to Power Load Forecasting

Yasser Al-Rashid, and Janet M. Twomey, Wichita State University, Wichita, KS

Experiments in Predicting Financial Time Series Using Modular Neural Network Architectures

Karsten Schierholt, and Cihan H. Dagli, University of Missouri-Rolla, Rolla, MO

An Artificial Neural Network Based Medium Term Electrical Load and Energy Forecast for a Rapidly Growing Utility

Saleh M. Al-Alawi, Sultan Qaboos University, Sultanate of Oman, OMAN; Syed M. Islam, The University of Newcastle, Newcastle, AUSTRALIA; Khaled A. Ellithy, University of Cairo, CAIRO, EGYPT; and Cihan H. Dagli, University of Missouri-Rolla, Rolla, MO

An Efficient Neural Network Model to Predict Stock Market by Using Conditional Density Estimation

Sabyasachi Ghoshray, Florida International University, Miami, FL

Artificial Neural Network in Time Series Analysis: An Ocean Engineering Experience

K. Ramanitharan, and C. W. Li, The Hong Kong Polytechnic University, Hyung Hon, Kowloon, HONG KONG

AUTHOR INDEX

A

Acar, L., WP1.1, WP2.1

Adair, K.L., WP2.3

Adhami, R.R., TP2.3, WP1.2

Aguirre, A.H., TA2.3

Ahmed, M.N., WA2.2

Akay, M., SP1.4

Akhmetov, D., MA2.2, WA2.4

Al-Alawi, S.M., WP2.4

Albers, D.J., TP2.1

Albin, Jr., D., WP2.1

Alexander, J.R., WP1.1

Allada, V., WP2.3

Al-Rashid, Y., WP2.4

Anand, S., TP2.2

Ananthraman, S., TP1.3

Anmala, J., TP2.4

Archuleta, M., MP1.4

Austin, J., WA2.3

B

Bakthula, N.R., WA2.2

Barrett, J.J., MP1.3

Barrett, L.K., WP1.2

Bartlett, E.B., MA2.1

Basheer, I.A., TP2.3, TP2.4

Bauer, K.W., TP1.3

Bécus, G.A., WP2.1

Bedford, D.F., WA2.3

Benedek, S., MP2.4

Biles, W., MP2.2

Bishop, J.B., TP1.3

Bock, P., WP1.2

Boddy, L., TP2.3

Bodine, H.L., WA2.3

Bodruzzaman, M., WA2.4

Book, R., MP1.3

Bounajem, E., TP2.4

Boustany, N.M., WP2.3

Boypati, S.V., WA2.2

Bradley, W.M.D., TA2.1

Branco, P.J.C., MP1.2

Bruce, J.W., WP1.2

Bruce, L.M., TP2.3, WP1.2

Buczak, A.L., MP1.3

Burghart, J.H., TA2.1

Bush, M., TA2.3

C

Cai, L.., MP1.3

Cantor, R.J., WA2.3, WP2.3

Cao, W., TA2.1

Carotta, M.C., TA2.4

Carter, J.N., TA2.3

Cha, J., TP1.2

Chacon M., M.I., TP1.2

Challoo, R., MA2.4, MP1.2

Chan, K.C., MP2.3

Chandrashekhara, K., WP1.1

Chang, C.A., MP1.4

Chang, T.S., MP2.1

Chellapilla, K., WP1.4

Chen, C.L.P., MA2.2, WP1.4

Chen, D., MP1.2

Chen, C.H., WP2.2

Chen, H.C., TP2.2

Chen, J., MP1.3

Chen, V.C.P., MP1.4

Chen, Y.T., TP1.4

Cheng, S., WP2.1

Cheng, T.C., TP2.4

Cheng, Y., WP2.1

Chiang, Chih-Tang, MA2.2

Chiang, Ching-Tsan, MA2.2,

MP2.1

Chinnam, R.B., MP2.4

Christiansen, A.D., TA2.3

Cistelecan, M., WP2.1

Coello, C.A.C., TA2.3

Cox, J.T., WP1.1

D

Dagli, C.H., TP1.1, TP2.2, WP2.4

Dai, L., WP2.2

Das, S., MP2.1

Dasarathy, B.V., TP2.1

Dasgupta, D., WA2.2

Dawes, R.L., WA2.3

Dechert, W.D., TP2.1

Dente, J.A., MP1.2

Dhiantravan, Y., TP2.1

Dimla Jr., D.E., TP1.4

Domínguez, I., WP1.1

Dote, Y., MA2.2, WA2.4

Dranger, T.S., TP2.1

Driankov, D., MP1.2

E

Eaton, P.H., WA2.1

Ehringer, W.D., WA2.2

Elder IV, J. F., SP1.2

El-Kwae, E.A., WA2.2

Ellithy, K.A., WP2.4

El-Sawah, M.S., WP2.1

Embrechts, M.J., MP2.4, TA2.4,

TP1.2

Enke, D., TP1.1

Enomoto, M., MP1.1

Erickson, K.T, WP1.1

Ersoy, O., SP1.3

Espinosa, I., WP1.1

Essawy, M.A., WA2.4

Ewald, H., TA2.2

F

Faglia, G., TA2.4

Fang, J.H., TP2.2

Farag, A.A., WA2.2

Fausett, L.V., TP1.2

Fell, J., TP1.3

Ferguson, R., TA2.4

Ferkinhoff, D.J., WP2.1

Ferroni, M., TA2.4

Fernández-Rodríguez, B., WA2.1

Feuring, T., TA2.4

Fialho, F.A.P., TA2.3

Fogel, D., TA1.1

Fodor, G., MP1.2

G

Ganesh, C., WP2.1

Garth, A.D.N., MP1.4

Gelenbe, E., Mon. Luncheon

Gero, J.S., MP2.3

Ghosh, J., SA1.3, MA2.1, TA2.1

Ghoshray, S., MP2.3, WP1.4,

WP2.4

Gisler, G.R., TP1.2

Glenn, J.F., WP2.1

Glover, R.J., WP1.4

Goel, S., TA2.4

Goldberg, D.E., MA2.3

Golden, B., MP1.4, WP2.2

Golden, R.M., WA2.3

Gong, K.F., WP2.1

Gordon, R.T., WA2.3, WP2.3

Govindaraju, R.S., TP2.4

Grantner, J.L., MP1.2

Graupe, D., TP1.3, TP2.3

Greene, K.A., TP1.3

Greenway, G.W., MP2.4

Grossberg, S., TP1.2

Grözinger, M., TP1.3

Guidi, V., TA2.4

Gunasekera, J.S., TP1.4

Gunasekera, M., TP1.4

Güngör, Z., WP2.2

H

Hajmeer, M.N., TP2.3

Hakura, J., MA2.4

Hall, E.L., TP2.2, MP2.2

Hammack, G.D., WA2.4

Hammitt, A., MA2.1

Han, S.S., MP1.3

Han, T., WP1.1

Haring, S., MP2.1, WP1.2

Harrier, D.J., WA2.3

Heathcock, R.B., WP1.2

Heller, M., WP1.4

Hemminger, T.L., WP1.3

Henke, G., TA2.4

Henley, S.S., WA2.3

Higginson, A.M., TA2.4, WP1.3

Hiller, J., WP1.2

Hong, C.M., MA2.2, MP2.1

Hong, J.E., MP2.4

Hooley, T., MP1.4

Hope, A.D., MP2.4

Hope, T., TP1.4

Hsieh, K.H., MP1.4

Hruska, S.I., WP2.3

Hu, L., TP2.4

Huang, C.Y., TA2.2

Hung, D.C.D., TA2.2

Huttunen, M., TP2.4

Hwang, J.N., MP2.1

Hwang, W.R., MP2.2

I

Ibrahim, E.T., WP2.1

Iokibe, T., Wed. Luncheon

Islam, S.M., WP2.4

J

Jaeger, R.C., MA2.2

Janikow, C.Z., MA2.3

Jaramillo-Mendez, N., WA2.1

Jasman, M.C., WP2.3

Jia, Z., TP1.4

Jin, J., WP1.2

Johnson, J.P., MA2.4

Jouse, W.C., TP1.1

K

Kabrisky, M., TP1.3

Kabuka, M.R., WA2.2

Kakazu, Y., MA2.1, MA2.4, MP1.3,

TA2.1, TP2.3

Kamel, E., MP2.2

Kasap, S., WP2.2

Kashner, T.M., WA2.3

Kazakov, V.A., MP2.3

Kelkar, N., MP2.2

Kelly III, W., MA2.4

Kercel, S. W., SA1.1

Keyes, C., TA2.3

Khedkar, P.S., TP1.4

Khiani, K.J., WA2.2

Khosla, P.K., MA2.3

Khotanzad, A., WA2.4

Kilmer, R.A., TP2.1

Klinkenberg, R.H., MP1.1

Kok, J.N., MP2.1, WP1.2

Kolodziej, C., MP1.2

Kordylewski, H., TP1.3, TP2.3

Kraft, R.P., TP1.2

KrishnaKumar, K., SA1.2, TP1.1,

WA2.1

Kubo, M., TP2.3

Kulkarni, A.D., WA2.2

Kumaralingam, R., TP1.1

Kuschewski, J.G., MP2.2

Kwon, S., TP2.4

L

Lahl, H., TA2.4

Lakshmanan, V., TA2.2

Laneman, J.N., WA2.1

Lara, R., WP1.1

Leavitt, K.W., MP2.3

Leighton, N.J., TP1.4

Levis, A.P., WP2.3

Li, C.W., WP2.4

Li, L.B., MP2.1

Lin, C.S., MA2.4

Lin, T.C., MP2.1

Lippe, W.M., TA2.4

Lister, P.M., TP1.4

Lo, J. T.H., TP1.1, WP1.3

Lu, Y., MA2.2

Mc

McLauchlan, R.A., MA2.4, MP1.2

M

Ma, Y.H., MA2.4

Maher, P.E., WP2.3

Marshall, A., MP2.4

Martinelli, G., TA2.4

Marudarajan, A.R., MP2.3, WP2.1

May, E.A., TP2.2

May, G.S., MP1.3

Mears, R.J., TA2.1

Meyer-Bäse, A., MA2.1

Millard, D.L., TP1.2

Miller, B.L., MA2.3

Miller, F.N., WA2.2

Miller, H., TP2.4

Minagawa, M., MP1.3

Mizuno, T., MA2.1

Moechtar, M., TP2.4

Monett, C.T., TP1.3

Morgan, G., WA2.3

Morris, C.W., TP2.3

Morrow, J.D., MA2.3

Mysore, R., TP2.4

N

Najjar, Y.M., TP2.3, TP2.4

Narendra, K. S., MA1.1

Neidhoefer, J., TP1.1, WA2.1

Nishimura, T., WA2.4

Nishita, K., TP1.1

Noel, S., WA2.4

Noroozi, S., MP2.4

Nouhi, A.N., MP2.3

O

Odrey, N.G., MA2.4

Olurotimi, O., MP2.1

Omar, S.I., MA2.4, MP1.2

Osman, K.A., TA2.4, WP1.3

Ozserim, S.B., TP2.1

P

Palaniswami, M., WP1.3

Pandya, N., WA2.1

Pao, Y., WA1.1

Patyra, M.J., MP1.2

Pauli, J., TP2.2

Peterson, G.E., WA2.1

Pomalaza-Raez, C.A., WP1.3

Pope, M.H., TP1.3

Poyntz-Wright, L., MP2.4

Prasad, N. R., MA2.2

Priemer, R., MP1.2, TP2.1

Prokhorov, D.V., WA2.1

Q

Quiza, J., WP1.1

R

Rabelo, L.C., TP1.4

Ragsdell, K.M., MP2.4

Ramagiri, N., MP1.4

Ramamurti, V., MA2.1

Ramanitharan, K., WP2.4

Rao, S.S., WP1.4

Raviwongse, R., WP2.3

Riley, H.N., MP2.3

Rine, D.C., MP1.3

Rivera, L., WP1.1

Roberts, S., TP1.2

Rogers, S.K., TP1.3

Rohatgi, A., MP1.3

Rollins, D.K., MP1.4

Rosandich, R.G., TP2.2

Röschke, J., TP1.3

Roulston, D.J., TA2.4

Rubin, M.A., TP1.2

Russell, C.A., TP1.3

S

Sabharwal, C.L., WA2.3

Sahan, N., WP2.1

Sahan, R.A., WP2.1

Salomon, R., MA2.3, MP1.1,

TA2.1

Samu, T., MP2.2

Sankaran, V., TP1.2

Santos, C.R.C., WP2.3

Sayeh, B., MP2.2

Sberveglieri, G., TA2.4

Schaina, L.M., WP2.1

Schierholt, K., WP2.4

Schwartz, C., TA2.3

Seifert, T.D., MP2.2

Shabani, F., MA2.2

Schäffner, C., TP1.3

Shamsi, A., WA2.4

Shi, H., MP1.1

Shi, H., MP2.4

Shi, J., WA2.4

Shoureshi, R., Banquet

Shrikhande, N., MP2.4

Silver-Warner, S.J., WP1.4

Smith, A.E., TP1.4, TP2.1

Smith, G., TP1.4

Smith, K., WP1.3

Smolleck, H. A., MA2.2

Smyser, C.P., WP1.1

Sprott, J.C., TP2.1

Srinivasa, L.N., TP2.2

St. Clair, D.C., MP1.1, WA2.3,

WP2.3

Steele, J.H.P., MP1.4, MP2.2

Stieper, M., TA2.2

Streilein, W.W., TP1.2

Sulung, MP2.3

Sun, X., MP1.4, WP2.2

Szpalski, M., TP1.3

T

Taha, I., TA2.1

Takita, M., MA2.1

Tantono, R., MP2.3

Teich, M. C., Tues. Luncheon

Tenhagen, A., TA2.4

Ter Weeme, M.P., TP1.4

Theiler, J., TP1.2

Thompson, E.A., WP2.1

Tite, C., MP2.4, TP1.4

Trafalis, T.B., TP1.1, WA2.4,

WP2.2

Turkbey, O., WP1.3, WP2.2

Tutunji, T.A., TP1.1

Twomey, J.M., TP1.4, WP2.4

U

Ueda, K., MP1.1

Uhl, T., TP1.3

Ukkonen, E., TP2.4

Ulug, E., TA2.2

V

Vaario, J., MP1.1

Van Bronkhorst, E., TA2.3

VanDerVeen, A., TP2.4

Van Loocke, P.R., MA2.1

Varadarajan, S., WP1.1

Vehviläinen, B., TP2.4

Viergever, M.A., MP2.1, WP1.2

Visnevski, N.A., WA2.1

Voyles, R.M., MA2.3

Vuwong, V., WP1.2

W

Wang, J., WP2.3

Wang, Q., WP1.4, WP2.2

Wasil, E., MP1.4

Whitrow, C., WP1.3

Wilamowski, B.M., MA2.2, MP2.1

Wilmes, E.J., WP1.1

Wilson, G.F., TP1.3

Wilson, L.S., WP1.2

Witt, A., TA2.2

Wittmaier, C., WA2.3

Wojakowski, J., TP2.4

Woo, P.Y., MP1.4

Woodley, R., WP1.1

Woyshville, M.J., WA2.3

Wright, A.L., TA2.2

Wu, C.Y., MA2.4

Wu, M.H., MP2.4

Wunsch II, D.C., WA2.1

X

Y

Yamany, S.M., WA2.2

Yang, B., TA2.4

Yang, G., TP1.4

Yang, H., TP2.2

Yatsuzuka, Y., MP1.1

Ye, S., WP1.4

Yokoi, H., MA2.1, MA2.4, TA2.1,

TP2.3

Yu, W., TA2.1

Yu, Z., MP1.2

Z

Zaman, M.M., WA2.4

Zein-Sabatto, S., MP2.2

Zhang, B., TP2.4

Zhao, Y., MP1.1

Zhou, Y., MP1.2

Zhu, J., MP1.4

Zhu, J.H., WA2.4

Zhuang, X., MP1.1

Zurada, J., TA2.2

ANNIE - 96 GUEST ROOM RESERVATION

November 10-13, 1996

Marriott's Pavilion Hotel

St. Louis, MO

Please type or print clearly in ink:

Dr./Mr./Ms.

_______________________________________________________________________________________________________

Last Name First Name M.I.

_______________________________________________________________________________________________________

Business or Organization

_______________________________________________________________________________________________________

Address (Home/Business)

_______________________________________________________________________________________________________

City State/Country Zip Code

__________________________________ _________________________ __________________________

Telephone (Business)

Arrival Date _____/_____/_____ Est. Arrival Time __________ (a.m., or p.m.) Departure Date_____/_____/_____

Please reserve ________________________________ rooms(s) for ____________________________________ people

To avoid duplication of reservations, please submit only one form when sharing accommodations with one or more individuals.

Name(s) of person(s) sharing accommodations _______________________________________________________________

Reservation request is subject to availability. There is a limited number of rooms set aside for your group.

Reservation requests will be accepted on a first come-first serve basis.

All rates are subject to state and local taxes. Cut off date for conference rate is October 21, 1996.

Guest rooms will be available for check in after 4:00 p.m., St. Louis time. Check out time is 12:00 noon.

Send all room reservations directly to Marriott, One Broadway St., St. Louis, MO 63102-1772.

DO NOT SEND HOTEL RESERVATIONS TO THE UNIVERSITY.

If paying by check, please include total of first and last night's deposit.

ROOM GUARANTEE: To guarantee your room throughout your stay, a non-refundable first night's deposit will be required. The deposit can be applied by a major credit card or check. Changes or cancellations after original reservation has been made should be made directly with hotel at (314) 421-1776.

Credit Card: ___ VISA ___ Discover ___ MasterCard ___ Amex ___ Diners Club Expiration Date:___________________

Card Number _______________________________ Signature: _______________________________________

ACCOMMODATIONS AND RATE (Please check) REQUESTED ROOM TYPE (Please check)

______ Single $82 _____ Triple $102 _____ Smoking _____ King Bed

1 Person 1 Bed 3 People 2 Beds _____ Non-smoking _____ 2 Double Beds

_____ No preference _____ No preference

_____ Double $82 _____ Quad $112

2 People 1 Bed 4 People 2 Beds

_____ Double/Double $92 _____ Parlors or Suites

2 People 2 Beds (Rates Available upon Request)

This registration form will also be used to guarantee for late arrival. Cancellations must be made by 6:00 P.M. St. Louis time.

ANNIE 1996 REGISTRATION FORM

November 10-13, 1996

Marriott's Pavilion Hotel

St. Louis, MO

Please type or print clearly in ink:

__________________________________ ________________________________________ __________

Last Name First Name M.I.

______________________________________________________________________________________________

Affiliation

______________________________________________________________________________________________

Business Address

____________________ ____________________ ____________________ ____________________ City State Zip Code Country

____________________ ____________________ ____________________ ____________________

Business Phone Fax Number E-mail Address WWW Address

Indicate Fee: Registration fee for at least one author MUST accompany your camera ready manuscript. (Fee includes reception, banquet, three luncheons and one copy of the proceedings).

$_____ $390 Author Registration

$_____ $430 On-Site Registration

$_____ $34 Extra Banquet Ticket (Spouse or Guest)

$_____ Tutorials $110 each (Circle Choices Below)

$_____ Other Fee (______________________________)

SUNDAY MORNING TUTORIAL SESSIONS

SA1.1 Wavelet Synthesis Methods

Dr. Stephen W. Kercel,

Oak Ridge National Lab, Oak Ridge, TN

SA1.2 Adaptive Critics: Theory and Applications

Dr. K. KrishnaKumar,

The University of Alabama, Tuscaloosa, AL

SA1.3 Advanced Pattern Recognition with Neural Networks

Dr. Joydeep Ghosh,

University of Texas at Austin, Austin, TX

$_____ $410 Advanced Registration

$_____ $285 Student Registration (Accompanied by a registered faculty member's name_____________________________)

$_____ $23 each Extra Luncheon Ticket (Please One)

_____ Monday _____ Tuesday _____ Wednesday

SUNDAY AFTERNOON TUTORIAL SESSIONS

SP1.2 Tools for Discovering Patterns in Data

Dr. John F. Elder IV, Elder Consulting and University of Virginia,

Charlottesville, VA

SP1.3 Learning of Statistical Neural Networks and Their

Applications

Dr. Okan Ersoy, Purdue University, West Lafayette, IN

SP1.4 Time-Frequency and Wavelet Methods with Engineering and Applications

Dr. Metin Akay, Rutgers University, Piscataway, NJ

TOTAL AMOUNT DUE: $ ______________ PAYABLE BY: (Please Indicate)

_____ Enclosed Check (Amount $_________ No. ___________)

(U.S. Currency, Payable to: University of Missouri - Rolla)

_____ Purchase Order No. _______________________________

Amount $________________________________________

_____ Credit Card:( ___ VISA ___ MasterCard ___ DISCOVER) Card Holder Name: __________________________________

Card Number: ____________________________________

Signature:_________________________________________

Expiration Date: _____________________________

SEND REGISTRATION TO: AUDIO VISUAL NEEDS: (Please Indicate Quantity)

ANNIE 1996 ______ Overhead Projector

Smart Engineering Systems Lab ______ 35mm Slide Projector

Dept. Of Engineering Management ______ Other(_______________________________________________)

University of Missouri-Rolla

1870 Miner Circle Phone: (573) 341-6576

Rolla, MO 65409-0370, USA Fax: (573) 341-6567

E-mail: rita@shuttle.cc.umr.edu, or register through Internet at http://www.umr.edu/~annie

PLEASE NOTE: All Hotel Reservation Forms should be sent directly to the Marriott Pavilion Hotel, not to the University. Thank You.