Abstract Aiming at the sensitivity of principal component analysis in dealing with the contaminated data and the property that its projection vectors are not sparse,a robust principal component analysis optimization model is proposed. The objective function of the proposed model adopts L1 norm and projective vectors are constrained by Lp norm. An iterative algorithm is used to solve the proposed model and the theoretical analysis shows that the algorithm can obtain the locally optimal solution. In addition,the kernel version is made by embedding kernel functions into the model. The experiments on UCI datasets and face datasets are performed to demonstrate the feasibility and effectiveness of the proposed method.
LIANG Zhi-Zheng,LI Yong,XIA Shi-Xiong等. Principal Component Analysis Based on L1-Norm Maximization with Lp-Norm Constraints[J]. , 2013, 26(2): 211-218.
[1]Jolliffe I T. Principal Component Analysis. 2nd Edition. New York,USA: Springer Verlag,2002 [2]Baccini A,Besse P,de Falguerolles A. A L1 Norm PCA and a Heuristic Approach // Proc of the International Conference on Ordinal and Symbolic Data Analysis. Paris,France,1995: 359-368 [3]Ding C,Zhou Ding,He Xiaofeng,et al. R1 PCA: Rotational Invariant L1 Norm Principal Component Analysis for Robust Subspace Factorization // Proc of the 23rd International Conference on Machine Learning. Pittsburgh,USA,2006: 281-288 [4]Nojun K. Principal Component Analysis Based on L1 Norm Maximization. IEEE Trans on Pattern Analysis and Machine Intelligence,2008,30(9): 1672-1680 [5]3Liang Zhizheng,Li Youfu. A Regularization Framework for Robust Dimensionality Reduction with Applications to Image Reconstruction and Feature Extraction. Pattern Recognition,2010,43(4): 1269-1281 [6]3Xu Huan,Caramanis C,Mannor S. Principal Component Analysis with Contaminated Data: The High Dimensional Case // Proc of the 23rd Annual Conference on Learning Theory. Haifa,Israel,2010: 490-502 [7]He Ran,Hu Baogang,Zheng Weishi,et al. Robust Principal Component Analysis Based on Maximum Correntropy Criterion. IEEE Trans on Image Processing,2011,20(6): 1485-1494 [8]Moghaddam B,Weiss Y,Avidan S. Spectral Bounds for Sparse PCA: Exact and Greedy Algorithms // Weiss Y,Schlkopf B,Platt J,eds. Advances in Neural Information Processing Systems. Cambridge,USA: MIT Press,2005,XV: 915-922 [9]Candes E J,Li X,Ma Y,et al. Robust Principal Component Analysis? Journal of ACM,2009,58(1): 1-37 [10]Horst R,Paradols P M,Thoai N V. Introduction to Global Optimization. 2nd Edition. Dordrecht,The Netherlands: Kluwer Academic Publisher,2000 [11]Freidman A. Foundations of Modern Analysis. New York,USA: Dover Publications,2010 [12]3Xu Zongben,Zhang Hai,Wang Yao,et al. L (1/2) Regularization. Science China: Information Sciences,2010,53(6): 1159-1169 [13]Schkopf B,Smola A J. Learning with Kernels. Cambridge,USA: MIT Press,2002 [14]Schkopf B,Smola A J,Müller K R. Nonlinear Component Analysis as a Kernel Eigenvalue Problem. Neural Computation,1998,10(5): 1299-1319 [15]Golub G H,van Loan F. Matrix Computations. 3rd Edition. Washington,USA: Johns Hopkins University Press,1996
2013, Vol. 26 
