The Ridge Function Representation of Polynomials and an Application to Neural Networks

The first goal of this paper is to establish some properties of the ridge function representation for multivariate polynomials, and the second one is to apply these results to the problem of approximation by neural networks. We find that for continuous functions, the rate of approximation obtained b...

Full description

Saved in:
Bibliographic Details
Published inActa mathematica Sinica. English series Vol. 27; no. 11; pp. 2169 - 2176
Main Authors Xie, Ting Fan, Cao, Fei Long
Format Journal Article
LanguageEnglish
Published Heidelberg Institute of Mathematics, Chinese Academy of Sciences and Chinese Mathematical Society 01.11.2011
Springer Nature B.V
Subjects
Algebra
Approximation
Mathematical analysis
Mathematics
Mathematics and Statistics
Multivariate analysis
Neural networks
Polynomials
Representations
Ridges
Studies
代数多项式
多元多项式
多项式函数
应用
神经网络
表示法
连续函数
逼近问题
approximation
Ridge function
41A63
neural network
41A05
polynomial
Online AccessGet full text

Cover

More Information
Summary:The first goal of this paper is to establish some properties of the ridge function representation for multivariate polynomials, and the second one is to apply these results to the problem of approximation by neural networks. We find that for continuous functions, the rate of approximation obtained by a neural network with one hidden layer is no slower than that of an algebraic polynomial.
Bibliography:The first goal of this paper is to establish some properties of the ridge function representation for multivariate polynomials, and the second one is to apply these results to the problem of approximation by neural networks. We find that for continuous functions, the rate of approximation obtained by a neural network with one hidden layer is no slower than that of an algebraic polynomial.
11-2039/O1
Ridge function, neural network, polynomial, approximation
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1439-8516
1439-7617
DOI:10.1007/s10114-011-9407-1