Title :
Digital hardware implementation of sigmoid function and its derivative for artificial neural networks
Author :
Gafsi, Z. ; Besbes, Khaoula
Abstract :
In this paper we propose a polynomial approximation of the sigmoid activation function and its derivative used in artificial neural networks, and we describe the design of the equivalent digital circuit using a floating-point representation for numbers. The simulation of the circuit realized with CMOS technology AMS 0.35μm under a frequency of 300 MHz shows the efficiency of the implementation.
Keywords :
CMOS digital integrated circuits; floating point arithmetic; integrated circuit design; neural chips; polynomial approximation; 0.35 micron; 300 MHz; CMOS technology; activation function; artificial neural networks; digital hardware implementation; equivalent digital circuit; floating-point representation; polynomial approximation; sigmoid function; Artificial neural networks; CMOS technology; Chebyshev approximation; Circuit simulation; Digital circuits; Frequency; Neural network hardware; Neurons; Polynomials; Silicon;
Conference_Titel :
Microelectronics, 2001. ICM 2001 Proceedings. The 13th International Conference on
Print_ISBN :
0-7803-7522-X
DOI :
10.1109/ICM.2001.997519
