Close Approximations of Sigmoid Functions by Sum of Steps for VLSI Implementation of Neural Networks

Valeriu Beiu; Jan A Peperstraete; Joos Vandewalle; Rudy Lawereins

Close Approximations of Sigmoid Functions by Sum of Steps for VLSI Implementation of Neural Networks

Valeriu Beiu, Jan A Peperstraete, Joos Vandewalle, Rudy Lawereins

Research output: Contribution to journal › Article › peer-review

Abstract

This paper is devoted to show that there are simple and accurate ways to compute a sigmoid nonlinearity and its derivative in digital hardware by sum of steps, and that threshold gate implementation of such algorithms are area-efficient when compared to other known methods.

Original language	English
Pages (from-to)	5-34
Journal	Scientific Annals of Computer Science "Alexandru Ioan Cuza" University of Iaşi
Volume	40
Publication status	Published - Jun 1 1994

Access to Document

http://www.info.uaic.ro/bin/Annals/Article?v=III&a=0

Cite this

@article{2521d48afa7c414c9b56113f8d15a770,

title = "Close Approximations of Sigmoid Functions by Sum of Steps for VLSI Implementation of Neural Networks",

abstract = "This paper is devoted to show that there are simple and accurate ways to compute a sigmoid nonlinearity and its derivative in digital hardware by sum of steps, and that threshold gate implementation of such algorithms are area-efficient when compared to other known methods.",

author = "Valeriu Beiu and Peperstraete, {Jan A} and Joos Vandewalle and Rudy Lawereins",

year = "1994",

month = jun,

day = "1",

language = "English",

volume = "40",

pages = "5--34",

journal = "Scientific Annals of Computer Science {"}Alexandru Ioan Cuza{"} University of Ia{\c s}i",

}

TY - JOUR

T1 - Close Approximations of Sigmoid Functions by Sum of Steps for VLSI Implementation of Neural Networks

AU - Beiu, Valeriu

AU - Peperstraete, Jan A

AU - Vandewalle, Joos

AU - Lawereins, Rudy

PY - 1994/6/1

Y1 - 1994/6/1

N2 - This paper is devoted to show that there are simple and accurate ways to compute a sigmoid nonlinearity and its derivative in digital hardware by sum of steps, and that threshold gate implementation of such algorithms are area-efficient when compared to other known methods.

AB - This paper is devoted to show that there are simple and accurate ways to compute a sigmoid nonlinearity and its derivative in digital hardware by sum of steps, and that threshold gate implementation of such algorithms are area-efficient when compared to other known methods.

M3 - Article

VL - 40

SP - 5

EP - 34

JO - Scientific Annals of Computer Science "Alexandru Ioan Cuza" University of Iaşi

JF - Scientific Annals of Computer Science "Alexandru Ioan Cuza" University of Iaşi

ER -

Close Approximations of Sigmoid Functions by Sum of Steps for VLSI Implementation of Neural Networks

Abstract

Access to Document

Fingerprint

Cite this