Abstract:In this paper, a new family of neural network (NN) operators is introduced. The idea is to consider a Durrmeyer-type version of the widely studied discrete NN operators by Costarelli and Spigler (Neural Netw 44:101–106, 2013). Such operators are constructed using special density functions generated from suitable sigmoidal functions, while the reconstruction coefficients are based on a convolution between a general kernel function and the function being reconstructed, f . Here, we investigate their approximation capabilities, establishing both pointwise and uniform convergence theorems for continuous functions. We also provide quantitative estimates for the approximation order thanks to the use of the modulus of continuity of f ; this turns out to be strongly influenced by the asymptotic behaviour of the sigmoidal function . Our study also shows that the estimates we provide are, under suitable assumptions, the best possible. Finally, -approximation is also established. At the end of the paper, examples of activation functions are discussed.

The New Approximation Operators with Sigmoidal Functions

Approximation by Exponential Type Neural Network Operators

Approximation by approximate interpolation neural networks with single hidden layer

Approximation Capability of a Novel Neural Network Model for Dynamic Systems

Approximation by a new kind of -Bernstein–Kantorovich operators

Approximation of neural network operators based on the Joukowski transformation

Double-quantitative $γ^{\ast}-$fuzzy coverings approximation operators

A Class of Interpolation Polynomial Operators and Approximation of Unbounded Continuous Functions

The Construction and Approximation of the Neural Network with Two Weights

Best Approximation and Inverse Results for Neural Network Operators

Approximation Rates for Neural Networks with General Activation Functions

Simultaneous approximation by neural network operators with applications to Voronovskaja formulas

Approximation theorems on graphs

A Novel Approximation Methodology and Its Efficient VLSI Implementation for the Sigmoid Function.

Neural Networks and the Best Trigomometric Approximation.

The approximation capabilities of Durrmeyer-type neural network operators

On Approximation of Nonbounded Continuous Functions

Approximation on Variable Exponent Spaces by Linear Integral Operators

An approximation by neural networkswith a fixed weight

Approximation Bounds by Neural Networks in L Ω P [-4Pt]

Approximation Characteristics for Diagonal Operators in Different Computational Settings