Advertisement

Robust PCA: Optimization of the Robust Reconstruction Error Over the Stiefel Manifold

  • Anastasia PodosinnikovaEmail author
  • Simon Setzer
  • Matthias Hein
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8753)

Abstract

It is well known that Principal Component Analysis (PCA) is strongly affected by outliers and a lot of effort has been put into robustification of PCA. In this paper we present a new algorithm for robust PCA minimizing the trimmed reconstruction error. By directly minimizing over the Stiefel manifold, we avoid deflation as often used by projection pursuit methods. In distinction to other methods for robust PCA, our method has no free parameter and is computationally very efficient. We illustrate the performance on various datasets including an application to background modeling and subtraction. Our method performs better or similar to current state-of-the-art methods while being faster.

Keywords

Principal Component Analysis Reconstruction Error Breakdown Point Robust Principal Component Analysis Ground Truth Information 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Notes

Acknowledgements

M.H. has been partially supported by the ERC Starting Grant NOLEPRO and M.H. and S.S. have been partially supported by the DFG Priority Program 1324, “Extraction of quantifiable information from complex systems”.

References

  1. 1.
    Bouwmans, T.: Recent advanced statistical background modeling for foreground detection: a systematic survey. Recent Pat. Comput. Sci. 4(3), 147–176 (2011)Google Scholar
  2. 2.
    Candès, E., Li, X., Ma, Y., Wright, J.: Robust principal component analysis? J. ACM 58(3), 11:1–11:37 (2011)CrossRefGoogle Scholar
  3. 3.
    Croux, C., Pilzmoser, P., Oliveira, M.R.: Algorithms for projection-pursuit robust principal component analysis. Chemometr. Intell. Lab. Syst. 87, 218–225 (2007)CrossRefGoogle Scholar
  4. 4.
    Hampel, F.R., Ronchetti, E.M., Rousseeuw, P.J., Stahel, W.A.: Robust Statistics: The Approach Based on Influence Functions. John Wiley and Sons, New York (1986)zbMATHGoogle Scholar
  5. 5.
    Higham, N.J., Schreiber, R.S.: Fast polar decomposition of an arbitrary matrix. SIAM J. Sci. Comput. 11(4), 648–655 (1990)MathSciNetCrossRefzbMATHGoogle Scholar
  6. 6.
    Horn, R., Johnson, C.: Matrix Analysis. Cambridge University Press, Cambridge (1990)zbMATHGoogle Scholar
  7. 7.
    Huber, P.J.: Projection pursuit. Ann. Stat. 13(2), 435–475 (1985)CrossRefzbMATHGoogle Scholar
  8. 8.
    Huber, P., Ronchetti, E.: Robust Statistics, 2nd edn. John Wiley and Sons, New York (2009)CrossRefzbMATHGoogle Scholar
  9. 9.
    Jolliffe, I.: Principal Component Analysis. Springer Series in Statistics, 2nd edn. Springer, New York (2002)zbMATHGoogle Scholar
  10. 10.
    Journée, M., Nesterov, Y., Richtárik, P., Sepulchre, R.: Generalized power method for sparse principal component analysis. J. Mach. Learn. Res. 1(1), 517–553 (2010)Google Scholar
  11. 11.
    Li, G., Chen, Z.: Projection-pursuit approach to robust dispersion matrices and principal components: primary theory and Monte Carlo. J. Am. Stat. Assoc. 80(391), 759–766 (1985)CrossRefzbMATHGoogle Scholar
  12. 12.
    Mackey, L.: Deflation methods for sparse PCA. In: 24th Conference on Neural Information Processing Systems, pp. 1017–1024 (2009)Google Scholar
  13. 13.
    Mateos, G., Giannakis, G.: Robust PCA as bilinear decomposition with outlier-sparsity regularization. IEEE Trans. Signal Process. 60(10), 5176–5190 (2012)MathSciNetCrossRefGoogle Scholar
  14. 14.
    McCoy, M., Tropp, J.A.: Two proposals for robust PCA using semidefinite programming. Electron. J. Stat. 5, 1123–1160 (2011)MathSciNetCrossRefzbMATHGoogle Scholar
  15. 15.
    Rousseeuw, P.J.: Least median of squares regression. J. Am. Stat. Assoc. 79(388), 871–880 (1984)MathSciNetCrossRefzbMATHGoogle Scholar
  16. 16.
  17. 17.
    De la Torre, F., Black, M.: Robust principal component analysis for computer vision. In: 8th IEEE International Conference on Computer Vision, pp. 362–369 (2001)Google Scholar
  18. 18.
    Wright, J., Peng, Y., Ma, Y., Ganesh, A., Rao, S.: Robust principal component analysis: exact recovery of corrupted low-rank matrices by convex optimization. In: 24th Conference on Neural Information Processing Systems, pp. 2080–2088 (2009)Google Scholar
  19. 19.
    Xu, H., Caramanis, C., Mannor, S.: Outlier-robust PCA: the high dimensional case. IEEE Trans. Inf. Theory 59(1), 546–572 (2013)MathSciNetCrossRefGoogle Scholar
  20. 20.
    Xu, H., Caramanis, C., Sanghavi, S.: Robust PCA via outlier pursuit. IEEE Trans. Inf. Theory 58(5), 3047–3064 (2012)MathSciNetCrossRefGoogle Scholar
  21. 21.
    Xu, L., Yuille, A.L.: Robust principal component analysis by self-organizing rules based on statistical physics approach. IEEE Trans. Neural Networks 6, 131–143 (1995)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Anastasia Podosinnikova
    • 1
    Email author
  • Simon Setzer
    • 2
  • Matthias Hein
    • 2
  1. 1.INRIA – Sierra Project-TeamÉcole Normale SupérieureParisFrance
  2. 2.Computer Science DepartmentSaarland UniversitySaarbrückenGermany

Personalised recommendations