Random approximants and neural networks (Q1914798): Difference between revisions
From MaRDI portal
Removed claims |
Changed an Item |
||
Property / author | |||
Property / author: Yu. I. Makovoz / rank | |||
Normal rank | |||
Property / reviewed by | |||
Property / reviewed by: Wolfgang W. Breckner / rank | |||
Normal rank |
Revision as of 09:19, 16 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Random approximants and neural networks |
scientific article |
Statements
Random approximants and neural networks (English)
0 references
9 September 1997
0 references
Let \(D\) be a set with a probability measure \(\mu\), \(\mu(D)=1\), and let \(K\) be a compact subset of \(L_q(D,\mu)\), \(1\leq q<\infty\). For \(f\in L_q\), \(n=1,2,\dots\), let \(\rho_n(f,K)=\inf|f-g_n|_q\), where the infimum is taken over all \(g_n\) of the form \(g_n= \sum^n_{i=1}a_i\phi_i\), with arbitrary \(\phi_i\in K\) and \(a_i\in R\). It is shown that for \(f\in \text{conv}(\pm K)\), under some mild restrictions, \(\rho_n(f,K)\leq C_q\varepsilon_n(K)n^{-1/2}\), where \(\varepsilon_n(K)\to 0\) as \(n\to\infty\). This fact is used to estimate the errors of certain neural net approximations. For the latter, also the lower estimates of errors are given.
0 references