TY - JOUR
ID -
TI - L_p_ Approximation by ReLU Neural Networks
AU - Eman Samir Bhaya
PY - 2020
VL - 6
IS - 4
SP -
EP -
JO - Karbala International Journal of Modern Science مجلة كربلاء العالمية للعلوم الحديثة
SN - 2405609X 24056103
AB -
We know that we can use the neural networks for the approximation of functions for many types of activation functions. Here, we treat only neural networks with simple and particular activation function called rectified linear units (ReLU). The main aim of this paper is to introduce a type of constructive universal approximation theorem and estimate the error of the universal approximation. We will obtain optimal approximation if we have a basis independent of the target function. We prove a type of Debao Chen's theorem for approximation.
ER -