A Novel Adaptive Kernel for the RBF Neural Networks

被引：42

作者：

Khan, Shujaat ^{[1
]}

Naseem, Imran ^{[2
,3
]}

Togneri, Roberto ^{[3
]}

Bennamoun, Mohammed ^{[4
]}

机构：

[1] Iqra Univ, Fac Engn Sci & Technol, Shaheed E Millat Rd Ext, Karachi 75500, Pakistan

[2] Karachi Inst Econ & Technol, Coll Engn, Karachi 75190, Pakistan

[3] Univ Western Australia, Sch Elect Elect & Comp Engn, 35 Stirling Highway, Crawley, WA 6009, Australia

[4] Univ Western Australia, Sch Comp Sci & Software Engn, 35 Stirling Highway, Crawley, WA 6009, Australia

来源：

CIRCUITS SYSTEMS AND SIGNAL PROCESSING | 2017年 / 36卷 / 04期

关键词：

Artificial neural networks; Radial basis function; Gaussian kernel; Support vector machine; Euclidean distance; Cosine distance; Kernel fusion; LEARNING ALGORITHM; CLASSIFICATION;

D O I：

10.1007/s00034-016-0375-7

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

In this paper, we propose a novel adaptive kernel for the radial basis function neural networks. The proposed kernel adaptively fuses the Euclidean and cosine distance measures to exploit the reciprocating properties of the two. The proposed framework dynamically adapts the weights of the participating kernels using the gradient descent method, thereby alleviating the need for predetermined weights. The proposed method is shown to outperform the manual fusion of the kernels on three major problems of estimation, namely nonlinear system identification, patter classification and function approximation.

引用

页码：1639 / 1653

页数：15

共 27 条

[1] A Novel Kernel for RBF Based Neural Networks [J].

Aftab, Wasim ;

Moinuddin, Muhammad ;

Shaikh, Muhammad Shafique .

ABSTRACT AND APPLIED ANALYSIS, 2014,

[2] A fast online learning algorithm of radial basis function network with locality sensitive hashing [J].

Ali S.H.A. ;

Fukase K. ;

Ozawa S. .

Evolving Systems, 2016, 7 (3) :173-186

[3]

[Anonymous], THESIS

[4] Meta-cognitive RBF Network and its Projection Based Learning algorithm for classification problems [J].

Babu, G. Sateesh ;

Suresh, S. .

APPLIED SOFT COMPUTING, 2013, 13 (01) :654-666

[5] Parameter selection algorithm with self adaptive growing neural network classifier for diagnosis issues [J].

Barakat, M. ;

Lefebvre, D. ;

Khalil, M. ;

Druaux, F. ;

Mustapha, O. .

INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2013, 4 (03) :217-233

[6] 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

[7]

Chen W., 2014, Recent Advances in Radial Basis Function Collocation Methods, DOI DOI 10.1007/978-3-642-39572-7

[8] RADIAL BASIS FUNCTION NEURAL-NETWORK FOR APPROXIMATION AND ESTIMATION OF NONLINEAR STOCHASTIC DYNAMIC-SYSTEMS [J].

ELANAYAR, S ;

SHIN, YC .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (04) :594-603

[9]

Er MJ, 2002, IEEE T NEURAL NETWOR, V13, P697, DOI 10.1109/TNN.2002.1000134

[10] BeeRBF: A bee-inspired data clustering approach to design RBF neural network classifiers [J].

Ferreira Cruz, Davila Patricia ;

Maia, Renato Dourado ;

da Silva, Leandro Augusto ;

de Castro, Leandro Nunes .

NEUROCOMPUTING, 2016, 172 :427-437

← 1 2 3 →