The construction and approximation of the neural network with two weights

DOI10.1155/2014/892653MaRDI QIDQ2336809zbMATH OpenOpenAlexDBLPWikidataFDO

Authors Zhiyong Quan, Zhengqiu Zhang

Publication date 19 November 2019

Published in Journal of Applied Mathematics (Search for Journal in Brave)

Copyright license Creative Commons Attribution 3.0 Unported

Full work available at URL https://doi.org/10.1155/2014/892653

zbMATH Keywords

approximationBochner-Riesz meanneural network operatorsigmoidal function

Mathematics Subject Classification ID

Algorithms for approximation of functions (65D15) Numerical algorithms for specific classes of architectures (65Y10) Networks and circuits as models of computation; circuit complexity (68Q06) Learning and adaptive systems in artificial intelligence (68T05)

Review text

Summary: The technique of approximate partition of unity, the way of Fourier series, and inequality technique are used to construct a neural network with two weights and with sigmoidal functions. Furthermore by using inequality technique, we prove that the neural network with two weights can more precisely approximate any nonlinear continuous function than BP neural network constructed in [\textit{Z. Chen} et al., Appl. Math., Ser. B (Engl. Ed.) 27, No. 1, 69--77 (2012; Zbl 1265.41025)].

Recommendations

• The construction and approximation of some neural network operators
• A formula for the approximation of functions by single hidden layer neural networks with weights from two straight lines
• Constructive approximation by neural networks with positive integer weights
• Neural-network approximation of functions of several variables
• Approximation by neural networks with weights varying on a finite set of directions

Cites work

• An Integral Upper Bound for Neural Network Approximation
• Approximation results for neural network operators activated by sigmoidal functions
• Constructive approximate interpolation by neural networks
• scientific article; zbMATH DE number 3367521 (Why is no real title available?)
• Multivariate neural network operators with sigmoidal activation functions
• New study on neural networks: the essential order of approximation
• The approximation operators with sigmoidal functions
• The construction and approximation of some neural network operators
• The essential order of approximation for neural networks
• The new approximation operators with sigmoidal functions

Cited in

(2)

• The construction and approximation of some neural network operators
• Constructive approximation by neural networks with positive integer weights

This page was built for publication: The construction and approximation of the neural network with two weights

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2336809)