Simple approximation of sigmoidal functions: realistic design of digital neural networks capable of learning
- 1 January 1991
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1505-1508 vol.3
- https://doi.org/10.1109/iscas.1991.176661
Abstract
Two different approaches to nonlinearity simplification in neural nets are presented. Both the solutions are based on approximation of the sigmoidal mapper often used in neural networks (extensions are being considered to allow approximation of a more general class of functions). In particular, a first solution yielding a very simple architecture, but involving discontinuous functions is presented; a second solution, slightly more complex, but based on a continuous function is then presented. This second solution has been successfully used in conjunction with the classical generalized delta rule algorithm.Keywords
This publication has 3 references indexed in Scilit:
- Design, fabrication and evaluation of a 5-inch wafer scale neural network LSI composed on 576 digital neuronsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1990
- Fast Generation of Neuro-ASICsPublished by Springer Nature ,1990
- Parallel Distributed ProcessingPublished by MIT Press ,1986