Shift-add neural architecture

This article is focused on implementation of artificial neural networks in hardware. We give an overview of the shift-add neural arithmetics, which provide a complete set of functions suitable for fast perceptron and RBF network implementations. The set consists of logarithm, exponent, multiplicatio...

Full description

Saved in:
Bibliographic Details
Published inICECS'99. Proceedings of ICECS '99. 6th IEEE International Conference on Electronics, Circuits and Systems (Cat. No.99EX357) Vol. 1; pp. 411 - 414 vol.1
Main Authors Skrbek, M., Snorek, M.
Format Conference Proceeding
LanguageEnglish
Published IEEE 1999
Subjects
Adders
Arithmetic
Artificial neural networks
Computer architecture
Computer networks
Gaussian processes
Linear approximation
Neural network hardware
Neural networks
Radial basis function networks
Online AccessGet full text

Cover

More Information
Summary:This article is focused on implementation of artificial neural networks in hardware. We give an overview of the shift-add neural arithmetics, which provide a complete set of functions suitable for fast perceptron and RBF network implementations. The set consists of logarithm, exponent, multiplication, square, square root, sigmoid-like and Gauss-like functions. All functions are linearly approximated to be easy implementable. Furthermore, we show the gate-level implementation of all functions provided by the shift-add arithmetics. Only adders and barrel shifters are necessary to accomplish all functions. The functions are optimized for very short propagation delay (a few nanoseconds). The shift-add architecture was evaluated by both software simulation and on-chip design. Results of the on-chip design are presented in the last section of this article.
ISBN:0780356829
9780780356825
DOI:10.1109/ICECS.1999.812310