Journal of the Korean Data and Information Science Society

The Korean Data and Information Science Society (한국데이터정보과학회)
권호 표지

Incremental Eigenspace Model Applied To Kernel Principal Component Analysis

  • Kim, Byung-Joo (School of Computer & Information Engineering Youngsan University)
  • Published : 2003.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

An incremental kernel principal component analysis(IKPCA) is proposed for the nonlinear feature extraction from the data. The problem of batch kernel principal component analysis(KPCA) is that the computation becomes prohibitive when the data set is large. Another problem is that, in order to update the eigenvectors with another data, the whole eigenvectors should be recomputed. IKPCA overcomes this problem by incrementally updating the eigenspace model. IKPCA is more efficient in memory requirement than a batch KPCA and can be easily improved by re-learning the data. In our experiments we show that IKPCA is comparable in performance to a batch KPCA for the classification problem on nonlinear data set.

Keywords

References

  1. Principal Component Neural Networks: Theory and Applications Diamantaras, K.I.;Kung, S.Y.
  2. British Machine Vision Conference v.1 Incremental eigenalysis for classification Hall, P.;Marshall, D.;Martin, R.
  3. PAMI v.22 no.9 Merging and splitting eigenspace models Hall, P.;Marshall, D.;Martin, R.
  4. AICHE Journal v.37 no.2 Nonlinear principal component analysis using autoassociative neural networks Kramer, M. A.
  5. Master's thesis, Technische Universitat Berlin November Kernalgorithmen zur nichtlinearen Signal verarbeitung in Merkmalsraumen Mika, S.
  6. Neural Computation v.10 no.5 Nonlinear component analysis as a kernel eigenvalue problem Scholkopf, B.;Smola, A.;Muller, K.R.
  7. IEEE Transactions on Neural Networks v.10 Input Space versus Feature Space in Kernel-based Methods Scholkopf, B.;Mika, S.;Burges, C.;Knirsch, P.;Muller, K.R.;Smola, A.
  8. Neural Computation v.11 no.2 Mixtures of probabilistic principal component analysers Tipping, M.E.;Bishop, C.M.
  9. Proc. ESANN Support classifier based on asymmetric kernel function Tsuda, K.
  10. Neural Processing Letters. v.9 Least squares support vector machine classifiers Suyken, J.A.K.;Vandewalle
  11. Statistical learning theory Vapnik, V. N.