Approximation by neural networks with a bounded number of nodes at each level

被引：38

作者：

Gripenberg, G ^{[1
]}

机构：

[1] Aalto Univ, Inst Math, FIN-02015 Espoo, Finland

来源：

JOURNAL OF APPROXIMATION THEORY | 2003年 / 122卷 / 02期

关键词：

approximation; neural; network; multilayer;

D O I：

10.1016/S0021-9045(03)00078-9

中图分类号：

O1 [数学];

学科分类号：

0701 ; 070101 ;

摘要：

it is shown that the general approximation property of feed-forward multilayer perceptron networks can be achieved in networks where the number of nodes in each layer is bounded, but the number of layers grows to infinity. This is the case provided the node function is twice continuously differentiable and not linear. (C) 2003 Elsevier Science (USA). All rights reserved.

引用

收藏

页码：260 / 266

页数：7

相关论文

共 10 条

[1] Error bounds for approximation with neural networks [J].

Burger, M ;

Neubauer, A .

JOURNAL OF APPROXIMATION THEORY, 2001, 112 (02) :235-250

[2] BACKPROPAGATION NEURAL NETS WITH ONE AND 2 HIDDEN LAYERS [J].

DEVILLIERS, J ;

BARNARD, E .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1993, 4 (01) :136-141

[3] Approximation of continuous functions of several variables by an arbitrary nonlinear, continuous function of one variable, linear functions, and their superpositions [J].

Gorban, AN .

APPLIED MATHEMATICS LETTERS, 1998, 11 (03) :45-49

[4] MULTILAYER FEEDFORWARD NETWORKS WITH A NONPOLYNOMIAL ACTIVATION FUNCTION CAN APPROXIMATE ANY FUNCTION [J].

LESHNO, M ;

LIN, VY ;

PINKUS, A ;

SCHOCKEN, S .

NEURAL NETWORKS, 1993, 6 (06) :861-867

[5]

Lorentz GG, 1996, Constructive approximation: Advanced problems, DOI New York

[6] Lower bounds for approximation by MLP neural networks [J].

Maiorov, V ;

Pinkus, A .

NEUROCOMPUTING, 1999, 25 (1-3) :81-91

[7] Uniform approximation by neural networks [J].

Makovoz, Y .

JOURNAL OF APPROXIMATION THEORY, 1998, 95 (02) :215-228

[8]

Pinkus A., 1999, Acta Numerica, V8, P143, DOI 10.1017/S0962492900002919

[9] Universal approximation using feedforward neural networks: A survey of some existing methods, and some new results [J].

Scarselli, F ;

Tsoi, AC .

NEURAL NETWORKS, 1998, 11 (01) :15-37

[10] Capabilities of a four-layered feedforward neural network: Four layers versus three [J].

Tamura, S ;

Tateishi, M .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1997, 8 (02) :251-255