Abstract
Principal Component Analysis (PCA) has been a cornerstone of data analysis for more than a century, with important applications across most fields of science and engineering. However, despite its many strengths, PCA is known to have a major drawback: it is very sensitive to the presence of outliers among the processed data. To counteract the impact of outliers in data analysis, researchers have been long working on robust modifications of PCA. One of the most successful (and promising) PCA alternatives is L1-PCA. L1-PCA relies on the L1-norm of the processed data and, thus, tames any outliers that may exist in the dataset. Experimental studies in various applications have shown that L1-PCA (i) attains similar performance to PCA when the processed data are outlier-free and (ii) maintains sturdy resistance against outliers when the processed data are corrupted. Thus, L1-PCA is expected to play a significant role in the big-data era, when large datasets are often outlier corrupted. In this chapter, we present the theoretical foundations of L1-PCA, optimal and state-of-the-art approximate algorithms for its implementation, and some numerical studies that demonstrate its favorable performance.
References
Barnett, V., Lewis, T.: Outliers in Statistical Data. Wiley, New York, NY (1994)
Bishop, C.M.: Pattern Recognition and Machine Learning. Springer, New York, NY (2006)
Brooks, J.P., Dulá, J.H.: The L1-norm best-fit hyperplane problem. Appl. Math. Lett. 26, 51–55 (2013)
Brooks, J.P., Dulá, J.H., Boone, E.L.: A pure L1-norm principal component analysis. J. Comput. Stat. Data Anal. 61, 83–98 (2013)
Candès, E.J., Li, X., Ma, Y., Wright, J.: Robust principal component analysis? J. ACM 58(3), 37 (2011)
Chamadia, S., Pados, D.A.: Optimal sparse l1-norm principal-component analysis. In: Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, pp. 2686–2690 (2017a)
Chamadia, S., Pados, D.A.: Outlier processing via l1-principal subspaces. In: Proceedings of Florida Artificial Intelligence Research Society (FLAIRS), Marco Island, FL, pp 508–513 (2017b)
Ding, C., Zhou, D., He, X., Zha, H.: \(r_1\)-PCA: Rotational invariant L1-norm principal component analysis for robust subspace factorization. In: Proceedings of International Conference on Machine Learning (ICML 2006), Pittsburgh, PA, pp. 281–288 (2006)
Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification, 2nd edn. Wiley, New York, NY (2001)
Eriksson, A., van den Hengel, A.: In: Efficient computation of robust low-rank matrix approximations in the presence of missing data using the L1 norm. In: Proceedings of IEEE Conference on Computer Vision Pattern Recognition (CVPR), San Francisco, CA, USA, pp. 771–778 (2010)
Golub, G.H.: Some modified matrix eigenvalue problems. SIAM Rev. 15, 318–334 (1973)
Johnson, M., Savakis, A.: Fast L1-eigenfaces for robust face recognition. In: Proceedings of IEEE Western New York Image Signal Processing Workshop (WNYISPW), Rochester, NY, pp. 1–5 (2014)
Jolliffe, I.T.: Principal Component Analysis. Springer, New York, NY (1986)
Ke, Q., Kanade, T.: Robust subspace computation using L1 norm. Techincal report Internal Technical Report, Computer Science Department, Carnegie Mellon University, Pittsburgh, PA, USA, CMU-CS-03-172 (2003)
Ke, Q., Kanade, T.: Robust L1 norm factorization in the presence of outliers and missing data by alternative convex programming. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), San Diego, CA, pp. 739–746 (2005)
Kundu, S., Markopoulos, PP., Pados, D.A.: Fast computation of the L1-principal component of real-valued data. In: Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Florence, Italy, pp. 8028–8032 (2014)
Kwak, N.: Principal component analysis based on L1-norm maximization. IEEE Trans. Patt. Anal. Mach. Intell. 30, 1672–1680 (2008)
Kwak, N., Oh, J.: Feature extraction for one-class classification problems: enhancements to biased discriminant analysis. Patt. Recog. 42, 17–26 (2009)
Liu, Y., Pados, D.A.: Compressed-sensed-domain L1-PCA video surveillance. IEEE Trans. Multimedia 18(3), 351–363 (2016)
Markopoulos, P.: Optimal algorithms for L1-norm principal component analysis: new tools for signal processing and machine learning with few and/or faulty training data. Ph.D. thesis, State University of New York at Buffalo (2015)
Markopoulos, P.P.: Reduced-rank filtering on L1-norm subspaces. In: Proceedings of IEEE Sensor Array Multichannel Signal Processing Workshop (SAM), Rio de Janeiro, Brazil, pp.1–5 (2016)
Markopoulos, P.P., Ahmad, F.: Indoor human motion classification by L1-norm subspaces of micro-doppler signatures. In: Proceedings of IEEE Radar Conference (Radarcon), Seattle, WA, pp. 1807–1810 (2017)
Markopoulos, P.P., Karystinos, G.N., Pados, D.A.: Some options for L1-subspace signal processing. In: Proceedings of 10th International Sympoisum on Wireless Communication System (ISWCS), Ilmenau, Germany, pp. 622–626 (2013)
Markopoulos, P.P., Karystinos, G.N., Pados, D.A.: Optimal algorithms for L1-subspace signal processing. IEEE Trans. Signal Process. 62, 5046–5058 (2014a)
Markopoulos, P.P., Tsagkarakis, N., Pados, D.A., Karystinos, G.N.: Direction finding with L1-norm subspaces. In: Proceedings of Commercial Sensing Conference on SPIE Defence Security Sensors (DSS), Baltimore MD, pp. 91-090J1–91-090J11 (2014b)
Markopoulos, P.P., Kundu, S., Pados, D.A.: L1-fusion: Robust linear-time image recovery from few severely corrupted copies. In: Proceedings of IEEE International Conference on Image Processing (ICIP), Quebec City, Canada, pp. 1225–1229 (2015)
Markopoulos, P.P., Kundu, S., Chamadia, S., Pados, D.A.: L1-norm principal-component analysis via bit flipping. In: Proceedings of IEEE International Conference on Machine Learning Applications (ICMLA), Anaheim, CA, pp. 326–332 (2016a)
Markopoulos, P.P., Tsagkarakis, N., Pados, D.A., Karystinos, G.N.: Direction-of-arrival estimation from l1-norm principal components. In: Proceedings of IEEE International Sympoisum on Phased Array Systems and Technology (PAST), Boston, MA, pp. 1–6 (2016b)
Markopoulos, P.P., Kundu, S., Chamadia, S., Pados, D.: Efficient l1-norm principal-component analysis via bit flipping. IEEE Transactions on Signal Processing (2016b)
McCoy, M., Tropp, J.A.: Two proposals for robust PCA using semidefinite programming. Electron. J. Stat. 5, 1123–1160 (2011)
Meng, D., Zhao, Q., Xu, Z.: Improve robustness of sparse PCA by L1-norm maximization. Patt. Recog. 45, 487–497 (2012)
Nie, F., Huang, H., Ding, C., Luo, D., Wang, H.: Robust principal component analysis with non-greedy L1-norm maximization. In: Proceedings of International Joint Conference on Artificial intelligence (IJCAI), Barcelona, Spain, pp. 1433–1438 (2011)
Pearson, K.: On lines and planes of closest fit to systems of points in space. Philos. Mag. 2, 559–572 (1901)
Schönemann, P.H.: A generalized solution of the orthogonal procrustes problem. Psychometrika 31(1), 1–10 (1966)
Tsagkarakis, N., Markopoulos, P.P., Pados, D.A.: Direction finding by complex L1-principal component analysis. In: Proceedings of IEEE 16th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), Stockholm, Sweden, pp. 475–479 (2015)
Tsagkarakis, N., Markopoulos, P.P., Pados, D.A.: On the l1-norm approximation of a matrix by another of lower rank. In: Proceedings of IEEE International Conference on Machine Learning Applications (IEEE ICMLA 2016), IEEE, pp. 768–773 (2016)
Wang, H.: Block principal component analysis with L1-norm for image analysis. Patt. Recog. Lett. 33, 537–542 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Markopoulos, P.P., Kundu, S., Chamadia, S., Tsagkarakis, N., Pados, D.A. (2018). Outlier-Resistant Data Processing with L1-Norm Principal Component Analysis. In: Naik, G. (eds) Advances in Principal Component Analysis. Springer, Singapore. https://doi.org/10.1007/978-981-10-6704-4_6
Download citation
DOI: https://doi.org/10.1007/978-981-10-6704-4_6
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-6703-7
Online ISBN: 978-981-10-6704-4
eBook Packages: EngineeringEngineering (R0)