Limitations and Future Trends in Neural Computation

by Ablameyko, Sergey; Gori, Marco; Goras, Liviu; Piuri, Vincenzo; NATO ADVANCED RESEARCH WORKSHOP ON LIMIT

• ISBN13: 9781586033248
• ISBN10: 1586033247
• Format: Hardcover
• Copyright: 2003-08-01
• Publisher: Ios Pr Inc

Summary

This book reports critical analyses on complexity issues in the continuum setting and on generalization to new examples, which are two basic milestones in learning from examples in connectionist models. The problem of loading the weights of neural networks, which is often framed as continuous optimization, has been the target of many criticisms, since the potential solution of any learning problem is severely limited by the presence of local minimal in the error function. The maturity of the field requires to convert the quest for a general solution to all learning problems into the understanding of which learning problems are likely to be solved efficiently. Likewise, the notion of efficient solution needs to be formalized so as to provide useful comparisons with the traditional theory of computational complexity in the discrete setting. The book covers these topics focussing also on recent developments in computational mathematics, where interesting notions of computational complexity emerge in the continuum setting.

Table of Contents

Preface	v
Chapter 1. Continuous Problem Solving and Computational Suspiciousness,
Marco Gori	1	(22)
Chapter 2. The Complexity of Computing with Continuous Time Devices,
Asa Ben-Hur, Hava Siegelmann and Shmuel Fishman	23	(22)
Chapter 3. Energy-Based Computation with Symmetric Hopfield Nets,
Jiri Sima	45	(26)
Chapter 4. Computational Complexity and the Elusiveness of Global Optima,
Marcello Pelillo	71	(24)
Chapter 5. Impact of Neural Networks on Signal Processing and Communications,
Simon Haykin	95	(20)
Chapter 6. From Clustering Data to Traveling as a Salesman: Empirical Risk Approximation as a Learning Theory,
Joachim M. Buhmann	115	(26)
Chapter 7. Learning High-dimensional Data,
Michel Verleysen	141	(22)
Chapter 8. The Curse of Dimensionality and the Blessing of Multiply Hybrid Networks,
Nathan Intrator	163	(14)
Chapter 9. A Kind of Neural Computing Goes Practical: TeraOPS, Stored Programmable, Analog-and-Logic Visual Microprocessors - a Review,
Tamas Roska	177	(8)
Chapter 10. On Pattern Formation in Cellular Neural Networks,
Liviu Goras	185	(22)
Chapter 11. Reliability of Man-System Interaction and Theory of Neural Networks Operations,
Mirko Novák and Zdenëk Votruba	207	(12)
Chapter 12. From Neural Networks to Intelligent Systems: Selected Aspects of Training, Application and Evolution,
Vladimir Golovko	219	(26)
Author Index	245