Abstract
The model for improving the robustness of sparse principal component analysis (PCA) is proposed in this paper. Instead of the l 2-norm variance utilized in the conventional sparse PCA model, the proposed model maximizes the l 1-norm variance, which is less sensitive to noise and outlier. To ensure sparsity, l p -norm (0 ⩽ p ⩽ 1) constraint, which is more general and effective than l 1-norm, is considered. A simple yet efficient algorithm is developed against the proposed model. The complexity of the algorithm approximately linearly increases with both of the size and the dimensionality of the given data, which is comparable to or better than the current sparse PCA methods. The proposed algorithm is also proved to converge to a reasonable local optimum of the model. The efficiency and robustness of the algorithm is verified by a series of experiments on both synthetic and digit number image data.
Similar content being viewed by others
References
Jollife I. Principal Component Analysis. New York: Springer-Verlag, 1986
Jollife I. Rotation of principal components: choice of normalization constraints. J Appl Stat, 1995, 22: 29–35
Cadima J, Jollife I. Loadings and correlations in the interpretation of principal components. J Appl Stat, 1995, 22: 203–214
Tibshirani R. Regression shrinkage and selection via the Lasso. J Roy Statist Soc Ser B Met, 1996, 58: 267–288
Jollife I, Uddin M. A modied principal component technique based on the Lasso. J Comput Graph Stat, 2003, 12: 531–547
Zou H, Hastie T, Tibshirani R. Sparse principal component analysis. J Comput Graph Stat, 2006, 15: 265–286
d’Aspremont A, El Ghaoui L, Jordan M I, et al. A direct formulation for sparse PCA using semidefinite programming. SIAM Rev, 2007, 49: 434–448
Shen H P, Huang J Z. Sparse principal component analysis via regularized low rank matrix approximation. J Multivariate Anal, 2008, 99: 1015–1034
Sigg C D, Buhmann J M. Expectation maximization for sparse and non-negative PCA. In: Proceedings of the 25th International Conference on Machine Learning, Helsinki, 2008. 960–967
Journée M, Nesterov Y, Richtárik P, et al. Generalized power method for sparse principal component analysis. J Mach Learn Res, 2010, 11: 517–55
Sriperumbudur B K, Torres D A, Lanckriet G R. Sparse eigen methods by D.C. programming. In: Proceedings of the 24th International Conference on Machine learning, Corvallis, 2007. 831–838
Lu Z S, Zhang Y. An augmented Lagrangian approach for sparse principal component analysis. Math Program, 2012, 135: 149–193.
Moghaddam B, Weiss Y, Avidan S. Spectral bounds for sparse PCA: exact and greedy algorithms. In: Proceedings of the 19th Conference on Neural Information Processing Systems, Vancouver, 2005. 915–922
d’Aspremont A, Bach F R, Ghaoui L E. Optimal solutions for sparse principal component analysis. J Mach Learn Res, 2008, 9: 1269–1294
Croux C, Filzmoser P, Fritz H. Robust sparse principal component analysis. Technometrics, 2013, 55: 202–214
Meng D Y, Zhao Q, Xu Z B. Improve robustness of sparse PCA based on L 1-norm maximization. Patt Recog, 2012, 45: 487–497
Kwak N. Principal component analysis based on L 1-norm maximization. IEEE Trans Patt Anal Mach Intell, 2008, 30: 1672–1680
De la Torre F, Black M J. A framework for robust subspace learning. Int J Comput Vis, 2003, 54: 117–142
Aanas H, Fisker R, Astrom K, et al. Robust factorization. IEEE Trans Patt Anal Mach Intell, 2002, 24: 1215–1225
Ding C, Zhou D, He X, et al. R1-PCA: rotational invariant L 1-norm principal component analysis for robust subspace factorization. In: Proceedings of the 23rd International Conference on Machine Learning, Pittsburgh, 2006. 281–288
Baccini A, Besse P, Falguerolles A D. A L 1-norm PCA and a heuristic approach. In: Diday E, Lechevalier Y, Opitz O, eds. Ordinal and Symbolic Data Analysis. New York: Springer-Verlag, 1996. 359–368
Ke Q, Kanade T. Robust L 1 norm factorization in the presence of outliers and missing data by alternative convex programming. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington D.C.: IEEE, 2005. 739–746
Marjanovic G, Solo V. On L q optimization and matrix completion. IEEE Trans Signal Process, 2012, 60: 5714–5724
Blumensath T, Davies M E. Iterative hard thresholding for compressed sensing. Appl Comput Harmonic Anal, 2009, 27: 265–274
Xu Z B, Chang X Y, Xu F M, et al. L 1/2-regularization: a thresholding representation theory and a fast solver. IEEE Trans Neural Netw Learn Syst, 2012, 23: 1013–1027
Xu Z B, Zhang H, Wang Y, et al. L 1/2-regularization. Sci China Inf Sci, 2010, 53: 1159–1169
Zeng J S, Fang J, Xu Z B. Sparse SAR imaging based on L 1/2 regularization. Sci China Inf Sci, 2012, 55: 1755–1775
Xu Z B, Guo H L, Wang Y, et al. Representative of L 1/2 regularization among L q (0 < q ⩽ 1) regularizations: an experimental study based on phase diagram. Acta Autom Sin, 2012, 38: 1225–1228
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Zhao, Q., Meng, D. & Xu, Z. Robust sparse principal component analysis. Sci. China Inf. Sci. 57, 1–14 (2014). https://doi.org/10.1007/s11432-013-4970-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11432-013-4970-y