Trigonometric Approximation and 2π-Periodic Neural Network Approximation

Abstract

Many articles studied best trigonometric approximation and many researchers worked on the neural network approximation, but no one related the best trigonometric approximation to neural network approximation. We define trigonometric activation function, then we use it to obtain neural network, which we use it as a best approximation for functions in spaces for . That what we shall introduce in our work have.