Skip to main content
SpringerLink
Log in

Optimizing widths with PSO for center selection of Gaussian radial basis function networks

  • Research Paper
  • Published:
Science China Information Sciences Aims and scope Submit manuscript

Abstract

The radial basis function (RBF) centers play different roles in determining the classification capability of a Gaussian radial basis function neural network (GRBFNN) and should hold different width values. However, it is very hard and time-consuming to optimize the centers and widths at the same time. In this paper, we introduce a new insight into this problem. We explore the impact of the definition of widths on the selection of the centers, propose an optimization algorithm of the RBF widths in order to select proper centers from the center candidate pool, and improve the classification performance of the GRBFNN. The design of the objective function of the optimization algorithm is based on the local mapping capability of each Gaussian RBF. Further, in the design of the objective function, we also handle the imbalanced problem which may occur even when different local regions have the same number of examples. Finally, the recursive orthogonal least square (ROLS) and genetic algorithm (GA), which are usually adopted to optimize the RBF centers, are separately used to select the centers from the center candidates with the initialized widths, in order to testify the validity of our proposed width initialization strategy on the selection of centers. Our experimental results show that, compared with the heuristic width setting method, the width optimization strategy makes the selected centers more appropriate, and improves the classification performance of the GRBFNN. Moreover, the GRBFNN constructed by our method can attain better classification performance than the RBF LS-SVM, which is a state-of-the-art classifier.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Huang Y Z, Fan N. Learning from interpolated images using neural networks for digital forensics. In: Proceedings of 23rd IEEE Conference on Computer Vision and Pattern Recognition, San Francisco, 2010. 177–182

    Google Scholar 

  2. Zhang D B, Hu D W, Shen L C, et al. Design of an artificial bionic neural network to control fish-robot’s locomotion. Neurocomputing, 2008, 71: 648–654

    Article  Google Scholar 

  3. Wang S T, Fu D, Xu M, et al. Advanced fuzzy cellular neural network: application to CT liver images. Artif Intell Med, 2007, 39: 65–77

    Article  Google Scholar 

  4. Moody J, Darken C J. Fast learning in network of locally-tuned processing units. Neural Comput, 1989, 1: 281–294

    Article  Google Scholar 

  5. Barceló-Rico F, Díez J L, Bondia J. N New possibilistic method for discovering linear local behavior using hyper-gaussian distributed membership function. K Knowl Inf Syst, 2012, 30: 377–403

    Article  Google Scholar 

  6. Uykan Z, Guzelis C. Input-output clustering for determining the centers of radial basis function networks. In: Proceedings of ECCTD, Budapest, 1997. 435–439

    Google Scholar 

  7. Gomm J B, Yu D L. Selecting radial basis function network centers with recursive orthogonal least squares training. IEEE Trans Neural Netw, 2000, 11: 306–314

    Article  Google Scholar 

  8. Zhou B D, Yao H L, Shi M H, et al. An new immune genetic algorithm based on uniform design sampling. Knowl Inf Syst, 2012, 31: 389–403

    Article  Google Scholar 

  9. Lacerda E, de Carvalho A, Ludermir T. Evolutionary optimization of RBF networks. Int J Neural Syst, 2003, 11: 287–294

    Article  Google Scholar 

  10. Hong X, Chen S. A new RBF neural network with boundary value constraints. IEEE Trans Syst Man Cybern B-Cybern, 2009, 39: 298–303

    Article  Google Scholar 

  11. Yeung D, Ng W, Wang D, et al. Localized generalization error model and its application to architecture selection for radial basic function neural network. IEEE Trans Neural Netw, 2007, 18: 1294–1305

    Article  Google Scholar 

  12. Huang D S, Du J X. A constructive hybrid structure optimization methodology for radial basis probabilistic neural networks. IEEE Trans Neural Networks, 2008, 19: 2099–2115

    Article  Google Scholar 

  13. Musavi M T, Ahmed W, Chan K H, et al. On the training of radial basis function classifiers. Neural Netw, 1992, 5: 595–603

    Article  Google Scholar 

  14. Joo Er M, Wu S, Lu J, et al. Face recognition with radial basis function (RBF) neural networks. IEEE Trans Neural Netw, 2002, 13: 697–710

    Article  Google Scholar 

  15. Nopiah Z M, Khairir M I, Abdullah S, et al. Time complexity estimation and optimisation of the genetic algorithm clustering method. WSEAS Trans Math, 2010, 9: 334–344

    MathSciNet  Google Scholar 

  16. Chen S, Wu Y, Luk B L. Combined genetic algorithm optimization and regularized orthogonal least squares learning for radial basis function networks. IEEE Trans Neural Netw, 1999, 10: 1239–1243

    Article  Google Scholar 

  17. Zhao Z Q, Huang D S, Sun B Y. Human face recognition based on multi-features using neural networks committee. Pattern Recognit Lett, 2004, 25: 1351–1358

    Article  Google Scholar 

  18. Fu X, Wang L. Data dimensionality reduction with application to simplifying RBF network structure and improving classification performance. IEEE Trans Syst Man Cybern B-Cybern, 2003, 33: 399–409

    Article  Google Scholar 

  19. Kennedy J, Eberhart R C. Swarm Intelligence. San Mateo: Morgan Kaufmann, 2001

    Google Scholar 

  20. Masafumi M, Mummadi V, Fuhito T, et al. Maximum power point tracking of multiple photovoltaic arrays: a PSO approach. IEEE Trans Aerosp Electron Syst, 2011, 47: 367–380

    Article  Google Scholar 

  21. Trelea I C. The particle swarm optimization algorithm: convergence analysis and parameter selection. Inf Process Lett, 2003, 85: 317–325

    Article  MATH  MathSciNet  Google Scholar 

  22. Bishop C M. Neural Networks for Pattern Recognition. New York: Oxford University Press, 1996

    MATH  Google Scholar 

  23. Hastie T, Tibshirani R. Discriminant analysis by Gaussian mixtures. J Roy Stat Soc Ser B-Stat Method, 1996, 58: 155–176

    MATH  MathSciNet  Google Scholar 

  24. Poggio T, Girosi F. Networks for approximation and learning. Proc IEEE, 1990, 78: 1481–1497

    Article  Google Scholar 

  25. Figueiredo M A T. On Gaussian radial basis function approximations: interpretation, extensions, and learning strategies. In: 15th International Conference on Pattern Recognition, Barcelona, 2000. 618–621

    Google Scholar 

  26. Costa G, Manco G, Ortale R, et al. From global to local and viceversa: uses of associative rule learning for classification in imprecise environments. Knowl Inf Syst, 2012, 33: 137–169

    Article  Google Scholar 

  27. Zhao Z Q. A novel modular neural network for imbalanced classification problems. Pattern Recognit Lett, 2009, 30: 783–788

    Article  Google Scholar 

  28. Zhou Z H, Liu X Y. Training cost-sensitive neural networks with methods addressing the class imbalance problem. IEEE Trans Knowl Data Eng, 2006, 18: 63–77

    Article  Google Scholar 

  29. Alejo R, Sotoca J M, Casan G A. An empirical study for the multi-class imbalance problem with neural networks. Lect Note Comput Sci, 2008, 5197: 479–486

    Article  Google Scholar 

  30. Majid A, Lee C H, Mahmood M T, et al. Impulse noise filtering based on noise-free pixels using genetic programming. Knowl Inf Syst, 2012, 32: 505–526

    Article  Google Scholar 

  31. McLachlan G, Peel D. Finite Mixture Models. New York: John Wiley & Sons, 2000

    Book  MATH  Google Scholar 

  32. Pernkopf F, Bouchaffra D. Genetic-aased EM algorithm for learning Gaussian mixture models. IEEE Trans Patt Anal Mach Intell, 2005, 27: 1344–1348

    Article  Google Scholar 

  33. Bajwa I S, Shahid N M, Asif M N, et al. Feature based image classification by using principal component analysis. J Graph Vision Image Process, 2009, 9: 11–17

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. College of Computer Science and Information Engineering, Hefei University of Technology, Hefei, 230009, China

    ZhongQiu Zhao, XinDong Wu & Jun Gao

  2. Department of Computer Science, University of Vermont, Burlington, Vermont, 05405, USA

    XinDong Wu

  3. Hefei Institute of Intelligent Machines, Chinese Academy of Sciences, Hefei, 230031, China

    CanYi Lu

  4. Department of Automation, University of Science and Technology of China, Hefei, 230026, China

    CanYi Lu

  5. Lab. Sciences de l’Information et des Systmes, University of Sud-Toulon Var, La Garde, R229-BP20132-83957, France

    Herve Glotin

Authors
  1. ZhongQiu Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. XinDong Wu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. CanYi Lu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Herve Glotin
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Jun Gao
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to XinDong Wu.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Zhao, Z., Wu, X., Lu, C. et al. Optimizing widths with PSO for center selection of Gaussian radial basis function networks. Sci. China Inf. Sci. 57, 1–17 (2014). https://doi.org/10.1007/s11432-013-4850-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11432-013-4850-5

Keywords

Search

Navigation