A novel LS-SVMs hyper-parameter selection based on particle swarm optimization

被引：145

作者：

Guo, X. C. ^{[1
,2
]}

Yang, J. H. ^{[1
]}

Wu, C. G. ^{[1
,3
]}

Wang, C. Y. ^{[1
]}

Liang, Y. C. ^{[1
]}

机构：

[1] Jilin Univ, Coll Comp Sci & Technol, Key Lab Symbol Computat & Knowledge Engn, Minist Educ, Changchun 130012, Peoples R China

[2] NE Dianli Univ, Coll Sci, Jilin 132012, Peoples R China

[3] Chinese Acad Sci, Natl Lab Pattern Recognit, Inst Automat, Beijing 100080, Peoples R China

来源：

NEUROCOMPUTING | 2008年 / 71卷 / 16-18期

关键词：

Least-squares support vector machines; Parameter selection; Particle swarm optimization; Classification;

D O I：

10.1016/j.neucom.2008.04.027

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The selection of hyper-parameters plays an important role to the performance of least-squares support vector machines (LS-SVMs). In this paper, a novel hyper-parameter selection method for LS-SVMs is presented based on the particle swarm optimization (PSO). The proposed method does not need any priori knowledge on the analytic property of the generalization performance measure and can be used to determine multiple hyper-parameters at the same time. The feasibility of this method is examined on benchmark data sets. Different kinds of kernel families are investigated by using the proposed method. Experimental results show that the best or quasi-best test performance could be obtained by using the scaling radial basis kernel function (SRBF) and RBF kernel functions, respectively. (C) 2008 Elsevier B.V. All rights reserved.

引用

页码：3211 / 3215

页数：5

共 43 条

[1] Fast cross-validation algorithms for least squares support vector machine and kernel ridge regression [J].

An, Senjian ;

Liu, Wanquan ;

Venkatesh, Svetha .

PATTERN RECOGNITION, 2007, 40 (08) :2154-2162

[2] Automatic model selection for the optimization of SVM kernels [J].

Ayat, NE ;

Cheriet, M ;

Suen, CY .

PATTERN RECOGNITION, 2005, 38 (10) :1733-1745

[3] Semisupervised PSO-SVM regression for biophysical parameter estimation [J].

Bazi, Yakoub ;

Melgani, Farid .

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2007, 45 (06) :1887-1895

[4] Leave-one-out bounds for support vector regression model selection [J].

Chang, MW ;

Lin, CJ .

NEURAL COMPUTATION, 2005, 17 (05) :1188-1222

[5]

Chapelle O, 2000, ADV NEUR IN, V12, P230

[6] Choosing multiple parameters for support vector machines [J].

Chapelle, O ;

Vapnik, V ;

Bousquet, O ;

Mukherjee, S .

MACHINE LEARNING, 2002, 46 (1-3) :131-159

[7]

CHRISTIANIMI N, 2000, INTRO SUPPORT VECTOR

[8] Efficient computations for large least square support vector machine classifiers [J].

Chua, KS .

PATTERN RECOGNITION LETTERS, 2003, 24 (1-3) :75-80

[9] Radius margin bounds for support vector machines with the RBF kernel [J].

Chung, KM ;

Kao, WC ;

Sun, CL ;

Wang, LL ;

Lin, CJ .

NEURAL COMPUTATION, 2003, 15 (11) :2643-2681

[10] The particle swarm - Explosion, stability, and convergence in a multidimensional complex space [J].

Clerc, M ;

Kennedy, J .

IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2002, 6 (01) :58-73

← 1 2 3 4 5 →