A fast multi-output RBF neural network construction method

被引：44
作者：
Du, Dajun [1 ,2 ]
Li, Kang [2 ]
Fei, Minrui [1 ]
机构：
[1] Shanghai Univ, Sch Mechatron Engn & Automat, Shanghai Key Lab Power Stn Automat Technol, Shanghai 200072, Peoples R China
[2] Queens Univ Belfast, Sch Elect Elect Engn & Comp Sci, Belfast BT9 5AH, Antrim, North Ireland
来源：
NEUROCOMPUTING | 2010年 / 73卷 / 10-12期
基金：
英国工程与自然科学研究理事会; 美国国家科学基金会;
关键词：
Linear-in-the-parameters model; Multi-output radial basis function (RBF) neural network; Center selection; Computational complexity analysis; Multi-output fast recursive algorithm (MFRA); LEAST-SQUARES ALGORITHM; FUNCTION APPROXIMATION; SYSTEM-IDENTIFICATION; LEARNING ALGORITHM; NONLINEAR-SYSTEMS; REGRESSION; SELECTION; CLASSIFICATION; DESIGN; SIZE;
D O I：
10.1016/j.neucom.2010.01.014
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
This paper investigates the center selection of multi-output radial basis function (RBF) networks, and a multi-output fast recursive algorithm (MFRA) is proposed. This method can not only reveal the significance of each candidate center based on the reduction in the trace of the error covariance matrix, but also can estimate the network weights simultaneously using a back substitution approach. The main contribution is that the center selection procedure and the weight estimation are performed within a well-defined regression context, leading to a significantly reduced computational complexity. The efficiency of the algorithm is confirmed by a computational complexity analysis, and simulation results demonstrate its effectiveness. (C) 2010 Elsevier B.V. All rights reserved.
引用
收藏
页码：2196 / 2202
页数：7
相关论文
共 36 条
[1]   IDENTIFICATION OF MIMO NON-LINEAR SYSTEMS USING A FORWARD-REGRESSION ORTHOGONAL ESTIMATOR [J].
BILLINGS, SA ;
CHEN, S ;
KORENBERG, MJ .
INTERNATIONAL JOURNAL OF CONTROL, 1989, 49 (06) :2157-2189
[2]   Sparse model identification using a forward orthogonal regression algorithm aided by mutual information [J].
Billings, Stephen A. ;
Wei, Hua-Liang .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2007, 18 (01) :306-310
[3]  
Broomhead D. S., 1988, Complex Systems, V2, P321
[4]   ORTHOGONAL LEAST-SQUARES METHODS AND THEIR APPLICATION TO NON-LINEAR SYSTEM-IDENTIFICATION [J].
CHEN, S ;
BILLINGS, SA ;
LUO, W .
INTERNATIONAL JOURNAL OF CONTROL, 1989, 50 (05) :1873-1896
[5]   Multi-output regression using a locally regularised orthogonal least-squares algorithm [J].
Chen, S .
IEE PROCEEDINGS-VISION IMAGE AND SIGNAL PROCESSING, 2002, 149 (04) :185-195
[6]   Orthogonal-least-squares regression: A unified approach for data modelling [J].
Chen, S. ;
Hong, X. ;
Luk, B. L. ;
Harris, C. J. .
NEUROCOMPUTING, 2009, 72 (10-12) :2670-2681
[7]   ORTHOGONAL LEAST-SQUARES LEARNING ALGORITHM FOR RADIAL BASIS FUNCTION NETWORKS [J].
CHEN, S ;
COWAN, CFN ;
GRANT, PM .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1991, 2 (02) :302-309
[8]   ORTHOGONAL LEAST-SQUARES ALGORITHM FOR TRAINING MULTIOUTPUT RADIAL BASIS FUNCTION NETWORKS [J].
CHEN, S ;
GRANT, PM ;
COWAN, CFN .
IEE PROCEEDINGS-F RADAR AND SIGNAL PROCESSING, 1992, 139 (06) :378-384
[9]   Construction of Tunable Radial Basis Function Networks Using Orthogonal Forward Selection [J].
Chen, Sheng ;
Hong, Xia ;
Luk, Bing L. ;
Harris, Chris J. .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2009, 39 (02) :457-466
[10]  
FISHER RA, 1988, UCI REPOSITORY MACHI
1234