An incremental multivariate regression method for function approximation from noisy data

被引：7

作者：

Carozza, M ^{[1
]}

Rampone, S ^{[1
]}

机构：

[1] Univ Sannio, INFM, Fac Sci MMFFNN, I-82100 Benevento, Italy

来源：

PATTERN RECOGNITION | 2001年 / 34卷 / 03期

关键词：

function approximation; noisy data; network size; genetic algorithm; generalization;

D O I：

10.1016/S0031-3203(00)00020-0

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this paper we consider the problem of approximating functions from noisy data. We propose an incremental supervised learning algorithm for RBF networks. Hidden Gaussian nodes are added in an iterative manner during the training process. For each new node added, the activation function center and the output connection weight are settled according to an extended chained version of the Nadaraja-Watson estimator. Then the variances of the activation functions are determined by an empirical risk-driven rule based on a genetic-like optimization technique. (C) 2001 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.

引用

页码：695 / 702

页数：8

共 27 条

[1]

[Anonymous], 1996, UMIACSTR9622

[2]

[Anonymous], 1992, MULTIVARIATE DENSITY

[3]

[Anonymous], 1990, SUPPORT VECTOR LEARN

[4]

Bishop C. M., 1995, NEURAL NETWORKS PATT

[5] A tutorial on Support Vector Machines for pattern recognition [J].

Burges, CJC .

DATA MINING AND KNOWLEDGE DISCOVERY, 1998, 2 (02) :121-167

[6]

CAROZZA M, IN PRESS INCREMENTAL

[7]

CHENTOUF R, 1996, AIHENP 96 EPFL UNIL

[8] MULTIVARIATE CLASSIFICATION THROUGH ADAPTIVE DELAUNAY-BASED C-0 SPLINE APPROXIMATION [J].

CUBANSKI, D ;

CYGANSKI, D .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1995, 17 (04) :403-417

[9]

ESPOSITO A, 1998, NEURAL NETS WIRN 98

[10]

Feller W., 1970, INTRO PROBABILITY TH, V1

← 1 2 3 →