This book approaches VLSI neural networks from a practical viewpoint, using case studies to show the full process of VLSI implementation of a network, and addressing the important issues of learning algorithms and limited precision effects. System aspects and low-power implementation issues a
Adaptive Analog VLSI Neural Systems
β Scribed by M. A. Jabri, R. J. Coggins, B. G. Flower (auth.)
- Publisher
- Springer Netherlands
- Year
- 1996
- Tongue
- English
- Leaves
- 261
- Edition
- 1
- Category
- Library
No coin nor oath required. For personal study only.
β¦ Synopsis
amplitude ~---. -----. -----. -----,-----. -----,-,~ VfT:jΒ·" 4. 50 4. 00 3. 50 q . 3. 00 /'\. ~ -'" : ! . 2. 50 ,: \ . . . 1! -. i "'" " 2. 00 1. 50 Β·Β·GOΒ·Β·O_O_ ,-. . . . &. , . ; D Q . " . . . / 1. 00 0. 50 0. 00 L. -----1. . ---. . l. -----:-:::''"::-::--::-::-'-:::-::------=--::-'-::-:=---=-=""=_:' 5. 00 10. 00 15. 00 Figure 7. 1 The morphology of ST and VT retrograde 1:1. Β© 1995 IEEE [Coggins, labri, Flower and Pickard {1995}]. ing to the analog domain. Additionally, the use of differential pair multipliers and current node summing in the network allows a minΒ imum of devices in the network itself and hence associated savings in power and area. However, in the last few decades analog signal processing has been used sparingly due to the effects of device offΒ sets, noise and drift*. The neural network architecture alleviates these problems to a large extent due to the fact that it is both highly parallel and adaptive. The fact that the network is trained to recognize morphologies with the analog circuits in-loop means that the synaptic weights can be adapted to cancel device offsets [Castro, Tam and Holler (1993); Castro and Sweet (1993)]. The impact of local un correlated noise is reduced by the parallelism of * Most fabrication processes have been optimised for digital design techniques which results in poor analog performance.
β¦ Table of Contents
Front Matter....Pages i-vii
Overview....Pages 1-5
Introduction to neural computing....Pages 7-16
MOS devices and circuits....Pages 17-56
Analog VLSI building blocks....Pages 57-88
Kakadu β a micropower neural network....Pages 89-103
Supervised learning in an analog framework....Pages 105-136
A micropower intracardiac electrogram classifier....Pages 137-156
On-chip perturbation based learning....Pages 157-169
An analog memory technique....Pages 171-179
Switched capacitor techniques....Pages 181-199
A high speed image understanding system....Pages 201-222
A Boltzmann learning system....Pages 223-247
Back Matter....Pages 249-262
β¦ Subjects
Computer System Implementation; Circuits and Systems; Processor Architectures; Electrical Engineering
π SIMILAR VOLUMES
<p>This volume contains the proceedings of a workshop on Analog Integrated Neural Systems held May 8, 1989, in connection with the International Symposium on Circuits and Systems. The presentations were chosen to encompass the entire range of topics currently under study in this exciting new discipl
<p><em>Cellular Neural Networks and Analog VLSI</em> brings together in one place important contributions and up-to-date research results in this fast moving area. <br/><em>Cellular Neural Networks and Analog VLSI</em> serves as an excellent reference, providing insight into some of the most challen
Analog Design for CMOS VLSI Systems is a comprehensive text that offers a detailed study of the background principles and the analog design techniques for CMOS-VLSI implementation. The book covers the physical operation and the modelling of MOS transistors. Discusses the key features of integrat