Return-Path: Received: from RI.CMU.EDU by A.GP.CS.CMU.EDU id aa14268; 14 Mar 94 19:25:31 EST Received: from netcom.netcom.com by RI.CMU.EDU id aa19768; 14 Mar 94 19:23:24 EST Received: from localhost by mail.netcom.com (8.6.4/SMI-4.1/Netcom) id QAA18621; Mon, 14 Mar 1994 16:24:16 -0800 Date: Mon, 14 Mar 94 16:24:14 PST From: IIS Corp To: mkant+fuzzy-faq@cs.cmu.edu Subject: A short course on fuzzy logic inference Message-ID: Sender: mkant@A.GP.CS.CMU.EDU Intelligent Inference Systems Corp. Presents Zadeh, Ruspini, Bezdek, Bonissone, and Berenji On Fuzzy Logic Inference Systems A Five Day Short Course Chicago, Illinois Seattle, Washington April 25-29, 1994 May 2-6, 1994 Introduced by Lotfi Zadeh, Fuzzy Logic methods may be used to design intelligent systems utilizing knowledge expressed in natural language. This methodology, an important source of artificial intelligence applications, permits the processing of both symbolic and numerical information. Fuzzy logic has been applied to control trains (Sendai subway), elevators, household appliances, cameras, and manufacturing processes. Systems designed and developed utilizing fuzzy-logic methods have been shown to be more efficient than those based on conventional approaches. In combination with Computational Neural Networks techniques, fuzzy-logic methods may be used to design robust adaptive control systems. This course discusses the application of fuzzy logic and neural networks techniques to the design of fuzzy and hybrid neuro-fuzzy systems. More than 11 case studies will be discussed in detail. At the completion of this course, you will have a full understanding of the benefits of this technology, will know about existing successful applications, and will develop the necessary understanding and knowledge to design and apply fuzzy logic to your particular needs. Who Should Attend: Engineers, technical managers and project leaders, scientists, systems analysts, as well as others who would like to have more knowledge about this emerging technology. About the course: This is the strongest and most complete course available on this topic. All the presenters of this course are pioneers of the field, including Professor Lotfi Zadeh, who introduced its seminal ideas and concepts. Even if you have already taken a course or tutorial on fuzzy logic, you should try to attend this course since it will provide you with a deeper understanding of the latest techniques and applications in the areas of fuzzy logic, soft computing, pattern recognition, intelligent control, computational neural networks, and adaptive neuro-fuzzy systems. Hands-on and Problem Solving Sessions: Include presentations by commercial tool developers on their available software and hardware products for fuzzy logic. ********************************************************** ** Spend 5 days with the pioneers of Fuzzy Logic ** ********************************************************** Instructors: Lotfi Zadeh, Ph.D. is the inventor and the "father of fuzzy logic". He has been on the faculty of Electrical Engineering departments at the Columbia University and University of California, Berkeley. He is now a Professor Emeritus and the director of the UC Berkeley's initiative on Soft Computing. He has won numerous awards including the Paul-Sabatier University Honorary Doctorate in 1986, Japan's Honda Award in 1989, IEEE Education Medal in 1973, IEEE Centennial Medal in 1984, and IEEE Richard W. Hamming Medal in 1992. Hamid R. Berenji, Ph.D. is a senior research scientist at the Artificial Intelligence Research Branch of NASA Ames Research Center in Moffett Field, California. He is the principal investigator of the research project on intelligent control, and was a program chairman for the IEEE International Conference on Neural Networks (ICNN'93) conference in San Francisco. He serves on the editorial board of several technical publications including as an associate editor for IEEE transactions on Fuzzy Systems and IEEE transactions for Neural Networks. He is a program cochairman for the 1994 IEEE conference on Fuzzy Systems, Orlando, Florida. Enrique Ruspini, Ph.D. is a senior computer scientist and a SRI Fellow at the Artificial Intelligence Center of SRI International in Menlo Park, CA. He has many years of experience in research in the theory and applications of fuzzy logic. He was the General Chairman of the IEEE International Conference on Fuzzy Systems (FUZZ- IEEE'93) and the IEEE International conference on Neural Networks (ICNN'93). He is a Fulbright Fellow, one of the founders of the North American Fuzzy Information Processing Society, and a recipient of that society's King-Sun Fu award. He is the Program Cochairman for the 1994 IEEE conference on Fuzzy Systems, Orlando, Florida. Jim Bezdek, Ph.D. currently holds an Eminent Scholar Chair with the Department of Computer Science at the U. of W. Florida. His research interests include pattern recognition, computational neural networks, image processing and machine vision, medical computing, and expert systems. He is the founding editor of the Int'l Jo. of Approximate Reasoning and the IEEE Trans. Fuzzy Systems; and is an associate editor of the: IEEE Trans. NN, and Int'l Journals of Applied Intelligence, General Systems, and Fuzzy Sets and Systems. He is a past president of IFSA (Int'l Fuzzy Systems Assoc.) and NAFIPS (North American Fuzzy Information Processing Society), and has been an ACM national lecturer for the 1990-93 program years. Dr. Bezdek is a fellow of the IEEE. Piero Bonissone, Ph.D. is a senior Computer Scientist with the Corporate Research and Development Center in Schenectady, NY, and an Adjunct Professor of Electrical, Computer and Systems Engineering at RPI. He has published numerous papers on approximate reasoning, fuzzy sets, pattern recognition, and expert systems. He is also a recipient of North American Fuzzy Information Processing Society's King-Sun Fu award. He was the program chairman of the FUZZ-IEEE'93 conference. He is the general chairman for the 1994 IEEE conference on Fuzzy Systems, Orlando, Florida. ***************************************************************** Course Certificate: Each student in the course will receive a personalized certificate indicating that he or she has completed this course taught by the pioneers of the field of fuzzy logic: Lotfi Zadeh, Enrique Ruspini, Jim Bezdek, Piero Bonissone, and Hamid Berenji Additional Information: Contact Intelligent Inference Systems Corp., Phone (408) 730-8345, Fax: (408) 730-8550 or send an electronic mail to: iiscorp@netcom.com Benefits: Learn how to design and apply fuzzy logic, understand the fundamentals of this field, learn about the most recent developments, learn about available software and hardware tools for fuzzy logic, and explore the wide range of applications of fuzzy logic (more than 11 case studies will be discussed) Intelligent Inference Systems (IIS Corp.) specializes in technical training and consulting in fuzzy logic, fuzzy control, neural networks, and knowledge-based systems. In-house courses are also offered. For further information on arranging an in-house course, contact: Intelligent Inference Systems Corp. P.O. Box 2908, Sunnyvale, CA 94087. Phone : (408) 730-8345 Fax: (408) 730-8550 email: iiscorp@netcom.com ************************************************ Course Outline Classes: 8:30 a.m. to 4:30 p.m. Hands-on and Problem Solving Sessions: 4:30 p.m. to 6:00 p.m. Day 1. Fundamentals of Fuzzy logic -- Enrique Ruspini * Fuzzy sets vs. crisp sets * Role of fuzzy sets in uncertainty management * Why fuzzy logic is needed? * Fuzzy logic vs. probability theory * Products based on fuzzy logic * Status of fuzzy computer chips * Calculus of If-Then rules * Approximate reasoning methods * Motivations for fuzzy logic * Fuzzy sets * Fuzzy set operations * Alternative combination operators * Fuzzy relations and mappings * The extension principle Fuzzy inferential methods * Representation of approximate rules * Generalized modus ponens * Possibility Theory * Translation rules * Understanding Fuzzy Logic * Possibility distribution as elastic constraints * Case Study 1: Mobile robot motion control * Case Study 2: Flexible arm manipulator Day 2: Advanced Fuzzy Logic and Intelligent Control -- Lotfi Zadeh, Hamid Berenji Taxonomy and interpretation of If-Then rules * Rules with exceptions and qualifications * Analysis of collections of Fuzzy If- Then Rules * Use of FA-Prolog * Algebraic operations on Fuzzy If- Then Rules * Computing with fuzzy probabilities * Induction of Fuzzy If-Then Rules from data * Relations with Neural Networks * Fundamentals of Intelligent Control * Artificial Intelligence and control Hierarchical control * Learning control systems * Fuzzy logic control * Designing a fuzzy logic controller * Knowledge Representation in fuzzy logic control * Fine-tuning a fuzzy logic controller * Case study 3: Cart-pole balancing * Case study 4: Fuzzy parking control * Applications of fuzzy logic control * Case study 5: Automatic train control * Case study 6: Helicopter control * Fuzzy logic hardwares and computer chips * Fuzzy logic software tools * Fuzzy system analysis * Fuzzy system identification * Structure identification of FLCs * Stability analysis of FLCs Day 3: Adaptive Fuzzy and Neural Network Systems -- Hamid Berenji Computational Neural Networks * Recurrent neural networks * CMAC architectures * Hybrid neural network and fuzzy logic controllers * Fuzzy logic control and backpropagation * Fuzzy logic control and reinforcement learning * Approximate Reasoning-based Intelligent Control (ARIC): * Single-layer neural networks (ARIC architecture) * Multi-layer neural networks (GARIC architecture) * Guiding reinforcement learning with fuzzy logic * Generating linear fuzzy rules from data using radial basis functions * Case study 7: GARIC applied to cart-pole balancing Case study 8: Space Shuttle attitude control with fuzzy logic and reinforcement learning Day 4: Approximate Reasoning in Knowledge-based systems -- Piero Bonissone Introduction to Knowledge Based Systems (KBS) * Topology of Approximate Reasoning Systems * Bayesian Network * Fuzziness in probabilistic systems * Practical Considerations for implementation * Dempster Shafer (Belief) Theory * Fuzziness in Dempster-Shafer Theory * Reasoning with uncertainty in rule based systems * PRIMO: A plausible reasoning system * Knowledge Representation * Multi-valued logics: Triangular norms and conorms * Control of Inference * Case study 9: Use of a Fuzzy Rule Based System in ASW Application * Software Engineering for KBS and FLC * Comparison of FLC with Classical Controller * A Software Perspective to FLCs * FLC Development Phase * Knowledge Representation * Compatibility Relations and Modus Ponens * Inference Process in FLCs * FLC Compilation and Run-time Phases * Case study 10: Example of Compilation and Run-time system for Fuzzy PI Day 5: Numerical Pattern Recognition -- Jim Bezdek Pattern Recognition * Object Data * Relational Data * Labeled Data * Clustering & Classification with Fuzzy Model * Partition Spaces * Case Study 11: Fuzzy c-Means * Applications Vignettes * Advanced Topics and Applications * Relational Clustering * Properties of * Fuzzy Relations * Fuzzy Similarity Relation Spaces * Decomposition of Transitive Closures * SAHN Clustering Algorithms * Convex Decompositions * Objective Function Approaches * Fuzzy Logic and Clustering Networks * Prototypes and Re-labeling in Clustering * Sequential Hard c-Means * Kohonen's LVQ and KSO Models * Generalized LVQ * Fuzzy LVQ * Fuzzy Logic and Classifier Networks * Biological Neural Models * Computational Neural Networks * Feed Forward Classifier Networks * Statistical Decision Theory General Information: IIS Corp. accepts registrations irrespective of race, creed, sex, color, physical handicap, and national or ethnic origin. This includes but it is not limited to admissions, employment, and educational services Registration Fee: $1395 Includes tuition, a full copy of the notes, and refreshments during the breaks $1295 Per person for teams of three or more from the same organization. Early registration: A $100 discount will be given to paid registrations received before February 28, 1994. A $75 processing fee is charged if registration is cancelled before March 25, 1994. No refund after March 25, 1994 but substitution is allowed at all times. Locations and Accommodations: Please arrange accommodation directly with the hotel. Special rates are available by mentioning "IIS Corp. Fuzzy Logic Course". For Chicago course: For Seattle Course: Embassy Suites Seattle Airport Hilton 707 E. Butterfield Rd. 17620 Pacific Hwy. South Lombard, IL 60148 Seattle, WA 98188 Telephone: (800) EMBASSY Telephone: (206) 244-4800 X------------------- (cut here) ----------------------- Registration Form Desired Location: ___ Chicago, IL ___ Seattle, WA Name: _________________________________________ Address: __________________________________________________ __________________________________________________ Business/Home Phone: _________________________ Fax:_________________ Course Fee: $1395 ($1295 per person for teams of three or more) ___ Check Enclosed ___ Money Order ___ Purchase Order ___ Billing authorization (enclosed) Credit Card payment: ___ Visa ___ Master Card ___ American Express Card #_________________ Expiration Date ____________________ Signature _____________________ Name on Card: (please print) ___________________________________ Please mail, fax, or email this form to: Intelligent Inference Systems Corp. P.O. Box 2908 Sunnyvale, CA 94087 Phone: (408) 730-8345 Fax: (408) 730-8550 email: iiscorp@netcom.com ______________________________________________________ Return-Path: Received: from RI.CMU.EDU by A.GP.CS.CMU.EDU id aa14243; 14 Mar 94 19:23:36 EST Received: from netcom.netcom.com by RI.CMU.EDU id aa19766; 14 Mar 94 19:22:53 EST Received: from localhost by mail.netcom.com (8.6.4/SMI-4.1/Netcom) id QAA18512; Mon, 14 Mar 1994 16:23:43 -0800 Date: Mon, 14 Mar 94 16:23:42 PST From: IIS Corp To: mkant+fuzzy-faq@cs.cmu.edu Subject: A short course on neural networks and genetic algorithms Message-ID: Sender: mkant@A.GP.CS.CMU.EDU Intelligent Inference Systems Corp. Presents Widrow, Rumelhart, Hammerstrom, and Koza On Neural Networks and Genetic Algorithms A Four Day Short Course San Francisco, CA April 18-21, 1994 Although early forms of neural networks have existed since the 1960's, advances in hardware, software, and neural algorithms have brought this technology to the fore for the 1990's. Neural networks can be trained and can self-learn to recognize patterns, to recognize speech, to predict weather from pressure patterns, and to perform control functions of considerable complexity. It is expected that this new technology will play an important role in the control of power systems, of industrial plants, of robotic systems, and in many other practical applications. They also represent an exciting opportunity for custom parallel architectures. Genetic algorithms are useful in solving many problems, including optimization problems in non- linear multi-dimensional spaces. Genetic programming extends the genetic algorithm to the domain of computer programs. In genetic programming, populations of programs are genetically bred to solve problems. Genetic programming can solve problems in system identification, classification, control, robotics, optimization, and pattern recognition. About this course: This course discusses the application of neural networks and genetic algorithms to the design of intelligent systems. Many case studies will be discussed in detail. At the completion of this course, you will have a full understanding of the benefits of neural networks and genetic algorithms, will know about existing successful applications, and will develop the necessary knowledge to design and apply neural network and genetic algorithms to your particular needs. Hands-on and Problem Solving Sessions: Include presentations by commercial tool developers on their available software and hardware products for neural network and genetic algorithms. Who Should Attend: Engineers, technical managers and project leaders, scientists, systems analysts, as well as others who are interested in the fields of neural networks and genetic algorithms. INSTRUCTORS: Bernard Widrow is a professor of electrical engineering at Stanford University. Before joining the Stanford faculty in 1959, he was with the Massachusetts Institute of Technology, Cambridge, Massachusetts. He is presently engaged in research and teaching in neural networks, pattern recognition, adaptive filtering, and adaptive control systems. He is associate editor of the journal "Adaptive Control and Signal Processing", "Neural Networks", and "Information Sciences". He is coauthor with S.D. Stearns of "Adaptive Signal Processing" (Prentice Hall). Dr. Widrow received the SB, SM, and ScD degrees from MIT in 1951, 1953, and 1956. He is a member of the American Association of University Professors, the Pattern Recognition Society, Sigma Xi, and Tau Beta PI. He is a fellow of the IEEE. He is past president of the International Neural Network society. dr. Widrow received the IEEE Alexander Graham Bell Medal in 1986 for exceptional contributions to the advancement of telecommunications. In 1991, he was designated by the IEEE Neural Networks Council as a pioneer in the field. David E. Rumelhart received his Ph.D. degree from Stanford University in Mathematical Psychology in 1967. He was an Assistant, Associate and a Full Professor in Psychology from 1967- 1987 at the University of California, San Diego, until becoming a Full Professor at Stanford University. While at San Diego he was a co- founder of the Institute of Cognitive Science. He has worked as a Cognitive Scientist, building computational models of human intelligence during most of this time. During the last 14 years Professor Rumelhart has concentrated his work on the development of "brain-style" or "neurally inspired" computational architectures. He is co-author of a two-volume work on this topic entitled Parallel Distributed Processing: Explorations in the Microstructure of Cognition, which has played an important role in the popularization of work on neural networks. He is a member of the honorary Society of Experimental Psychologists, a charter member of the American Association of Artificial Intelligence, and a recipient of a MacArthur Foundation Fellowship for his work on cognitive modeling. He is a charter member of the American Association for Artificial Intelligence, a Fellow of the American Association for the Advancement of Science, a Fellow of the American Academy of Arts, and a Fellow of the National Academy of Sciences. Dan Hammerstrom received the B.S. degree in Electrical Engineering, with distinction, from Montana State University, the M.S. degree in Electrical Engineering from Stanford University, and the Ph.D. degree in Electrical Engineering from the University of Illinois. He was on the faculty of Cornell University from 1977 to 1980 as an Assistant Professor. From 1980 to 1985 he worked for Intel where he participated in the development and implementation of the iAPX- 432 and i960, and, as a consultant, the iWarp systolic processor that was jointly developed by Intel and Carnegie Mellon University. He is an Associate Professor at the Oregon Graduate Institute where he is pursuing research in massively parallel VLSI architectures, and is the founder and Chief Technical Officer of Adaptive Solutions, Inc. He is the architect of the Adaptive Solutions CNAPS neurocomputer. He has been a Visiting Professor at the Royal Institute of Technology in Stockholm, Sweden. Dr. Hammerstrom's research interests are in the area of the VLSI architectures for pattern recognition. He has been an Associate Editor for the Journal of the International Neural Network Society, IEEE Transactions on Neural Networks, and the International Journal of Neural Networks. John R. Koza is a consulting professor in the computer science department at Stanford University. He is author of the 1992 book Genetic Programming: On the Programming of Computers by Means of Natural Selection from the MIT Press and a 1994 book Genetic Programming II: Scalable Automatic Programming by Means of Automatically Defined Functions from the MIT Press. He received his Ph.D in Computer Science from the University of Michigan in the field of machine learning and induction in 1972. Between 1973 and 1987 he was chief executive officer of Scientific Games Incorporated in Atlanta, and he is currently president of Third Millennium Venture Capital Limited in California. ***************************************************************** Course Certificate: Each student in the course will receive a personalized certificate indicating that he or she has completed this course taught by the pioneers of the field of Neural Networks and Genetic Algorithms: Widrow, Rumelhart, Hammerstrom, and Koza. Additional Information: Contact Intelligent Inference Systems Corp., Phone (408) 730-8345, Fax: (408) 730-8550 or send an electronic mail to: iiscorp@netcom.com Benefits: Learn how to design and apply neural networks and genetic algorithms, understand the fundamentals of this field, learn about the most recent developments, learn about available software and hardware tools for neural networks and genetic algorithms, and explore the wide range of their applications. Intelligent Inference Systems (IIS Corp.) specializes in technical training and consulting in fuzzy logic, fuzzy control, neural networks, genetic algorithms, and knowledge-based systems. In-house courses are also offered. For further information on arranging an in-house course, contact: Intelligent Inference Systems Corp. P.O. Box 2908, Sunnyvale, CA 94087. Phone : (408) 730-8345 Fax: (408) 730-8550 email: iiscorp@netcom.com ************************************************ Course Outline Classes: 8:30 a.m. to 4:30 p.m. Hands-on and Problem Solving Sessions: 4:30 p.m. to 6:00 p.m. Daily Topics: Day 1: Adaptive Signal Processing and Neural Networks --- Bernard Widrow Adaptive Signal Processing * Adaptive filters * Minimization of mean square error * Convergence and learning speed prediction * Noise cancellation * Deconvolution * Adaptive beamforming * Case study: Applications in telecommunications, channel equalization and echo cancelling * Case study: Applications in control systems, inverse control and cancellation of plant disturbance * Case study: Applications in Acoustics, active control of sound and vibration * Adaptive Neural Networks * Backpropagation * Complex information processing * Pattern recognition * Case study: truck backer-upper * Case Study: Weather forecasting Day 2: Brain-style Computation --- David Rumelhart Neural networks theory * Computational modeling * Neural information processing * Abstraction of computational neuroscience principles * Formalization of neuroscience principles * Conditions under which neural networks might be useful * General theory behind neural networks * Why do they work? * How to design and build them for particular applications * Theoretical methods in Cognitive Science * Parallel Distributed Processing * Foundations of Cognitive Science * Cognition and thought * Conditions for useful applications * Case Study: Speech recognition * Case study: Optical Character Recognition * Case study: Cursive handwriting * Case study: Medical Diagnosis Day 3: Neural Network/Pattern Recognition Hardware and Applications --- Dan Hammerstrom Introduction, Motivation, and Overview * Parallel Computer Architectures for Neural Networks * Why not use existing machines, sequential or parallel? * Some experiences * Amdhal's Law * Traditional Processors vs. Non-traditional structures * Case study: ICSI RAO * Flexibility vs. Performance * Case study: Intel/Nestor Ni1000 * Case study: Simens Synapse-1 * Case study: Adaptive Solutions CNAPS * Case study: ICSI CNS-1 * Case study: HNC SNAP Analog * Case study: CalTech Silicon Retina * Case study: Synaptics MICR Code Reader * Case study: Applications, OCR, Control, Financial Day 4: Genetic Algorithms --- John Koza Introduction to Genetic Algorithms * Representation schemes * Schema theorem * Optimization problems in non-linear multi- dimensional spacess * Darwinian fitness proportionate reproduction and crossover * Case study: multi-dimensional control * Case study: Pattern recognition * Case study: Optimization * Case study: Modeling and robotics * Simulated Annealing * Examples of applications * Genetic Programming in system identification, classification, control, robotics, and pattern recognition * Computer implementation * Available software * Sources of additional information * Evolution strategies * Evolutionary programming * Genetic programming * Automatic function definition in genetic programming * Genetic programming applications General Information: Registration Fee: $1,095 Includes tuition, a full copy of the notes, and refreshments during the breaks. $100 discount per person for teams of three or more from the same organization. Save an additional $100 when registering by March 18, 1994. A $75 processing fee is charged if registration is cancelled before April 1, 1994. No refund after April 1, 1994 but substitution is allowed at all times. Locations and Accommodations: Please arrange accommodation directly with the hotel. Special rates are available by mentioning IIS Corp. Short Course. For San Francisco Course: Westin Hotel 1 Old Bayshore Highway Milbrae, CA 94030 Tel: (415) 692-3500 X------------------- (cut here) ----------------------- Registration Form Name: _________________________________________ Address: __________________________________________________ __________________________________________________ Business/Home Phone: _________________________ Fax:_________________ Course Fee: $1095 ($995 per person for teams of three or more) ___ Check Enclosed ___ Money Order ___ Purchase Order ___ Billing authorization (enclosed) Credit Card payment: ___ Visa ___ Master Card ___ American Express Card #_________________ Expiration Date ____________________ Signature _____________________ Name on Card: (please print) ___________________________________ Please mail, fax, or email this form to: Intelligent Inference Systems Corp. P.O. Box 2908 Sunnyvale, CA 94087 Phone: (408) 730-8345 Fax: (408) 730-8550 email: iiscorp@netcom.com ______________________________________________________