Adaptive Resonance Theory Microchips: Circuit Design Techniques / Edition 1

by Teresa Serrano-Gotarredona, Bernabï Linares-Barranco, Andreas G. Andreou

Overview

Adaptive Resonance Theory Microchips describes circuit strategies resulting in efficient and functional adaptive resonance theory (ART) hardware systems. While ART algorithms have been developed in software by their creators, this is the first book that addresses efficient VLSI design of ART systems. All systems described in the book have been designed and fabricated (or are nearing completion) as VLSI microchips in anticipation of the impending proliferation of ART applications to autonomous intelligent systems. To accommodate these systems, the book not only provides circuit design techniques, but also validates them through experimental measurements. The book also includes a chapter tutorially describing four ART architectures (ART1, ARTMAP, Fuzzy-ART and Fuzzy-ARTMAP) while providing easily understandable MATLAB code examples to implement these four algorithms in software. In addition, an entire chapter is devoted to other potential applications for real-time data clustering and category learning.

Product Details

ISBN-13:	9780792382317
Publisher:	Springer US
Publication date:	08/31/1998
Series:	The Springer International Series in Engineering and Computer Science , #456
Edition description:	1998
Pages:	234
Product dimensions:	6.10(w) x 9.25(h) x 0.36(d)

Table of Contents

1. Adaptive Resonance Theory Algorithms.- 1.1 Introduction.- 1.2 ART1.- 1.3 ARTMAP.- 1.4 Fuzzy-ART.- 1.5 Fuzzy-ARTMAP.- 2. A Vlsi-Friendly ART1 Algorithm.- 2.1 The Modified ART1 Algorithm.- 2.2 Functional Differences between Original and Modified Model.- 3. ART1 And ARTMAP Vlsi Circuit Implementation.- 3.1 Introduction.- 3.2 Hardware-Oriented Attractive Properties of the ART1 Algorithm.- 3.3 Circuit Description.- 3.4 Modular System Expansivity.- 3.5 Implementation of Synaptic Current Sources.- 3.6 Experimental Results of First Prototype.- 3.7 Experimental Results of Second Prototype.- 4. A Current-Mode Multi-Chip WTA-MAX Circuit.- 4.1 Introduction.- 4.2 Operation Principle.- 4.3 Circuit Implementation.- 4.4 System Stability Coarse Analysis.- 4.5 System Stability Fine Analysis.- 4.6 Experimental Results.- 5. An ART1/ARTMAP/Fuzzy-ART/Fuzzy-ARTMAP Chip.- 5.1 The Synaptic Cell.- 5.2 Peripheral Cells.- 5.3 Concluding Remarks.- 6. Analog Learning Fuzzy Art Chips.- 6.1 Introduction.- 6.2 Summary of the Fuzzy-ART Algorithm.- 6.3 Current-Mode Fuzzy-ART Chip.- 6.4 Fuzzy-ART/VQ Chip.- 6.5 Conclusions.- 7. Some Potential Applications For Art Microchips.- 7.1 Portable Non-invasive Device for Determination of Concentrations of Biological Substances.- 7.2 Cardiac Arrhythmia Classifier for Implantable Pacemaker.- 7.3 Vehicle Interior Monitoring Device for Auto Alarm.- 7.4 Concluding Remarks.- Appendices.- A- MATLAB Codes for Adaptive Resonance Theory Algorithms.- A.1 MATLAB Code Example for ART1.- A.2 MATLAB Code Example for ARTMAP.- A.3 MATLAB Code Example for Fuzzy-ART.- A.4 MATLAB Code Example for Fuzzy-ARTMAP.- A. 5 Auxiliary Functions.- B- Computational Equivalence of the Original ART1 and the Modified ART1m Models.- B. l Direct Access to Subset and Superset Patterns.- B. 2 DirectAccess by Perfectly Learned Patterns (Theorem 1 of original ART1).- B. 3 Stable Choices in STM (Theorem 2 of original ART1).- B. 4 Initial Filter Values determine Search Order (Theorem 3 of original ART1).- B. 5 Learning on a Single Trial (Theorem 4 of original ART1.- B. 6 Stable Category Learning (Theorem 5 of original ART1.- B. 7 Direct Access after Learning Self-Stabilizes (Theorem 6 of original ART1).- B.8 Search Order(Theorem 7 of original ART1).- B.9 Biasing the Network towards Uncommitted Nodes.- B.10 Expanding Proofs to Fuzzy-ART.- B. 11 Remarks.- C- Systematic Width-and-Length Dependent CMOS Transistor Mismatch Characterization.- C.1 Mismatch Characterization Chip.- C.2 Mismatch Parameter Extraction and Statistical Characterization.- C.3 Characterization Results.- References.

Editorial Reviews

Describes circuit strategies resulting in efficient and functional adaptive resonance theory (ART) hardware systems, focusing on VLSI design of ART systems. All systems described have been designed and fabricated as VLSI microchips in anticipation of the impending proliferation of ART applications to autonomous intelligent systems. Provides circuit design techniques and demonstrates how they validate experimental measurements, and offers a tutorial on four ART architectures, accompanied by MATLAB code examples to implement these four algorithms in software. Annotation c. by Book News, Inc., Portland, Or.

Booknews