Sigmoid generators for neural computing using piecewise approximations

Author

Zhang, Ming ; Vassiliadis, Stamatis ; Delgado-Frias, José G.

Author_Institution

AT&T Wireless Services Inc., Kirkland, WA, USA

Volume

45

Issue

9

fYear

1996

fDate

9/1/1996 12:00:00 AM

Firstpage

1045

Lastpage

1049

Abstract

A piecewise second order approximation scheme is proposed for computing the sigmoid function. The scheme provides high performance with low implementation cost; thus, it is suitable for hardwired cost effective neural emulators. It is shown that an implementation of the sigmoid generator outperforms, in both precision and speed, existing schemes using a bit serial pipelined implementation. The proposed generator requires one multiplication, no look-up table and no addition. It has been estimated that the sigmoid output is generated with a maximum computation delay of 21 bit serial machine cycles representing a speedup of 1.57 to 2.23 over other proposals

Keywords

function generators; neural net architecture; neural nets; piecewise polynomial techniques; hardwired cost effective neural emulators; maximum computation delay; neural computing; piecewise approximations; second order approximation scheme; sigmoid generators; Computer networks; Costs; Delay estimation; High performance computing; Neural network hardware; Neural networks; Piecewise linear approximation; Proposals; Signal generators; Table lookup;

fLanguage

English

Journal_Title

Computers, IEEE Transactions on

Publisher

ieee

ISSN

0018-9340

Type

jour

DOI

10.1109/12.537127

Filename

537127