On sigmoidal functions
According to \textit{Chen Debao} [J. Approximation Theory Appl. 9, No. 3, 17-28 (1993; Zbl 0784.41011)] the function \(\sigma: \mathbb R\to \mathbb R\) is called sigmoidal, if \(\lim_{t\to+\infty}\sigma(t)=1\), \(\lim_{t\to-\infty}\sigma(t)=0\). The author discusses the uniform approximation of the function continuous on \([0,1]\) by the superpositions of infinitely smooth sigmoidal functions. It is shown that the best approximation by the sum of \(n\) superpositions does not exceed the best approximation by the polynomials of the \(n\)-th power. An example is considered which shows that the function smoothness is not determined by the approximation order of smallness.
- Approximation order to a function in \(\overline C({\mathbb R})\) by superposition of a sigmoidal function
- Degree of multivariate approximation by superposition of a sigmoidal function
- scientific article; zbMATH DE number 495819
- An analog of the Jackson-Nikol'skij theorem on the approximation by superpositions of sigmoidal functions
- An approximation by neural networks with a fixed weight
- Approximation of functions of finite variation by superpositions of a sigmoidal function.
- Sigmoidal theory
- Some applications of Sigmoid functions
- Sigmoidal cosine series on the interval
- SIGMOIDAL-TYPE SERIES EXPANSION
- Parameter Estimation of Sigmoid Superpositions: Dynamical System Approach
- Degree of multivariate approximation by superposition of a sigmoidal function
- Constructive approximation by neural networks with positive integer weights
- On the approximation of the cut and step functions by logistic and Gompertz functions
- scientific article; zbMATH DE number 7540043 (Why is no real title available?)
- Sigmoid functions for the smooth approximation to the absolute value function
This page was built for publication: On sigmoidal functions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5936679)
