New method of RKHS model complexity reduction

详细信息    查看全文

• 作者：Nadia Souilem ; Ilyes Elaissi…
• 关键词：RKHS ; Kernel method ; RKCCA ; KCCA ; PCA ; PT326 trainer
• 刊名：The International Journal of Advanced Manufacturing Technology
• 出版年：2016
• 出版时间：February 2016
• 年：2016
• 卷：82
• 期：5-8
• 页码：961-971
• 全文大小：730 KB
• 参考文献：1.Akaho S (2001) A kernel method for canonical correlation analysis. In: Proceedings of International Meeting on Psychometric Society (IMPS2001)
  2.Aronszajn N (1950) Theory of reproducing kernels. Trans Am Math Soc 68(3):337–404CrossRef MathSciNet MATH
  3.Bach FR, Jordan MI (2002) Kernel independent component analysis. J Mach Learn Res 3:1–48MathSciNet MATH
  4.Bretherton CS, Smith C, Wallace JM (1992) An intercomparison of methods for finding coupled patterns in climate data. J Clim 5:541–560CrossRef
  5.Cristianini N, Shawe-Taylor J (2000) An introduction to support vector machines and other kernel-based learning methods. Cambridge University Press, CambridgeCrossRef MATH
  6.Cristianini N, Shawe-Taylor J, Lodhi H (2002) Latent semantic kernels. J Intell Inf Syst 18(2/3):127–152, Special issue on automated text categorization CrossRef
  7.Elaissi I, Jaffal I, Taouali O, Messaoud H (2013) Online prediction model based SVD-KPCA methods. ISA Trans 52:96–104CrossRef
  8.Elaissi I, Messaoud H (2005) Predictive control of a drying loop modeled on RKHS. J Heat Transf Res 36(3):151–169CrossRef
  9.Elaissi I, Taouali O, Messaoud H (2010) Dimensionality reduction for supervised learning with reproducing kernel hilbert space. International Review of Automatic Control 3(2)
  10.Hardoon DR et al (2004) Canonical correlations analysis: an overview with application to learning methods. Neural Comput 16(12):2639–2664CrossRef MATH
  11.Hotelling H (1936) Relations between two sets of variates. Biometrika 28(3/4):321–377CrossRef MATH
  12.Ljung L (1987) System identification. Theory for the user. Prentice Hall
  13.Melzer T, Reiter M, Bischof H (2001) Nonlinear feature extraction using generalized canonical correlation analysis. In Proceedings of International Conference on Artificial Neural Networks (ICANN), pages 353–360
  14.Mercer J (1909) Functions of positive and negative type and their connection with the theory of integral equations. Philos Trans R Soc 209:415–446CrossRef MATH
  15.Nina O (2012) Restricted kernel canonical correlation analysis. LINEAR ALGEBRA AND ITS APPLICATIONS. 437(1), pp.1–13. JUL 1, 2012
  16.Saidi N, Taouali O, Messaoud H (2009) Modelling of a SISO and MIMO non linear communication channel using two modelling techniques. WSEAS Trans Circ Syst 8(I6):518–527
  17.Scholkopf B, Smola A, Muller K-R (1998) Nonlinear component analysis as kernel eigenvalue problem. Neural Comput 10:1299–1319CrossRef
  18.Souilem N, Elaissi I, Taouali O, Messaoud H (2013) New online RKPCA-RN kernel method applied to Tennessee Eastman process. WSEAS Transactions on Systems 12(7)
  19.Taouali O, Elaissi I, Messaoud H (2010) Online identification of nonlinear system using reduced kernel principal component analysis. Neural Computing and Applications 19
  20.Taouali O, Elaissi I, Messaoud H (2012) Online prediction model based on reduced kernel principal component analysis. Neural Comput Appl J 21(1):161–169CrossRef
  21.Taouali O, Elaissi I, Messaoud H (2012) Design and comparative study of online kernel methods identification of nonlinear system in RKHS space. Artif Intell Rev 37:289–300CrossRef
  22.Vapnik VN (1995) The nature of statistical learning theory. Springer, New YorkCrossRef MATH
  23.Vapnik VN (1998) Statistical learning theory. Wiley, New YorkMATH
  24.Wahba G (2000) An introduction to model building with reproducing kernel Hilbert spaces. Technical report No 1020. (Department of Statistics, University of Wisconsin-Madison)
  
• 作者单位：Nadia Souilem (1)
  Ilyes Elaissi (1)
  Hassani Messaoud (1)
  
  1. Laboratory of Automatic Signal and Image Processing, National School of Engineers of Monastir, University of Monastir, Monastir, 5019, Tunisia
  
• 刊物类别：Engineering
• 刊物主题：Industrial and Production Engineering
  Production and Logistics
  Mechanical Engineering
  Computer-Aided Engineering and Design
  
• 出版者：Springer London
• ISSN：1433-3015

文摘

In this paper, we propose a new kernel method with complexity reduction of reproducing kernel Hilbert space (RKHS) models. In RKHS models, the number of parameters is equal to the training set size; this leads to a complex representation. We propose a new method, the reduced kernel canonical correlation analysis (RKCCA), to reduce the number of parameters of RKHS models. This method consists on approximating the canonical correlation coefficients by a set of observation data. The proposed method is used to identify experimentally two nonlinear systems.