Abstract
Recently, some matrix exponential-based discriminant analysis methods were proposed for high dimensionality reduction. It has been shown that they often have more discriminant power than the corresponding discriminant analysis methods. However, one has to solve some large-scale matrix exponential eigenvalue problems which constitutes the bottleneck in this type of methods. The main contribution of this paper is twofold: First, we propose a framework of fast implementation on general matrix exponential-based discriminant analysis methods. The key is to equivalently transform large-scale matrix computation problems into much smaller ones. On the other hand, it was mentioned in Wang et al. (IEEE Trans Image Process 23:920–930, 2014) that the exponential model is more reliable than the original one and suppresses the sensitivity to pertubations. However, the interpretation is only heuristic, and to the best of our knowledge, there is no theoretical justification for reliability and stability of the matrix discriminant analysis methods. To fill in this gap, the second contribution of our work is to provide stability analysis for the fast exponential discriminant analysis method from a theoretical point of view. Numerical experiments illustrate the numerical behavior of the proposed algorithm, and demonstrate that our algorithm is more stable than many state-of-the-art algorithms for high dimensionality reduction.
Similar content being viewed by others
References
Ahmed, N.: Exponential discriminant regularization using nonnegative constraint and image descriptor. In: IEEE 9th international conference on emerging technologies, pp. 1–6 (2013)
Belhumeur, P., Hespanha, J., Kriegman, D.: Eigenfaces vs Fisherface: recognition using class-specific linear projection. IEEE Trans. Pattern Anal. Mach. Intell. 19, 711–720 (1997)
Chang, X., Paige, C.C., Stewart, G.W.: Preturbation analysis for the QR factorization. SIAM J. Matrix Anal. Appl. 18, 775–791 (1997)
Chen, H., Chang, H., Liu, T.: Local discriminant embedding and its variants. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 846–853 (2005)
Chen, L., Liao, H., Ko, M., Lin, J., Yu, G.: A new LDA-based face recognition system which can solve the small sample size problem. Pattern Recognit. 33, 1713–1726 (2000)
Chu, D., Thye, G.: A new and fast implementation for null space based linear discriminant analysis. Pattern Recognit. 43, 1373–1379 (2010)
Cover, T., Hart, P.: Nearest neighbor pattern classification. IEEE Trans. Inf. Theory 13, 21–27 (1967)
Dornaika, F., Bosaghzadeh, A.: Exponential local discriminant embedding and its application to face recognition. IEEE Trans. Syst. Man Cybern. Part B Cybern. 43, 921–934 (2013)
Dornaika, F., El Traboulsi, Y.: Matrix exponential based semi-supervised discriminant embedding for image classification. Pattern Recognit. 61, 92–103 (2017)
Duda, R., Hart, P., Stock, D.: Pattern Classification, 2nd edn. Wiley, New York (2004)
Friedman, J.: Regularized discriminant analysis. J. Am. Stat. Assoc. 84, 165–175 (1989)
Fukunaga, K.: Introduction to Statistical Pattern Recognition, 2nd edn. Elsevier Academic Press, London (1999)
Golub, G.H., Van Loan, C.F.: Matrix Computations, 4th edn. The Johns Hopkins university Press, Baltimore (2013)
Higham, N.J.: Functions of Matrices: Theory and Computation. SIAM, Philadelphia (2008)
He, J., Ding, L., Cui, M., Hu, Q.: Marginal Fisher analysis based on matrix exponential transformation. Chin. J. Comput. 10, 2196–2205 (2014). (in Chinese)
He, X., Niyogi, P.: Locality preserving projections. In: Thrun, S., Saul, L., Schölkopf, B. (eds.) Advances in Neural Information Processing Systems, vol. 16, pp. 153–160. MIT Press, Cambridge (2004)
He, X., Yan, S.C., Hu, Y., Niyogi, P., Zhang, H.J.: Face recognition using laplacianfaces. IEEE Trans. Pattern Anal. Mach. Intell. 27, 328–340 (2005)
Howland, P., Park, H.: Generalizing discriminant analysis using the generalized singular value decomposition. IEEE Trans. Pattern Anal. Mach. Intell. 26, 995–1006 (2004)
Jolliffe, I.: Principal Component Analysis. Springer, New York (1986)
Kokiopoulou, E., Chen, J., Saad, Y.: Trace optimization and eigenproblems in dimension reduction methods. Numer. Linear Algebra Appl. 18, 565–602 (2011)
Kokiopoulou, E., Saad, Y.: Orthogonal neighborhood preserving projections. In: Houston, T.X., Han, J., et al. (eds.) IEEE 5th International Conference on Data Mining (ICDM05), pp. 234–241. IEEE, New York (2005)
Kokiopoulou, E., Saad, Y.: Orthogonal neighborhood preserving projections: a projection-based dimensionality reduction technique. IEEE Trans. Pattern Anal. Mach. Intell. 29, 2143–2156 (2007)
Kramer, R., Young, A., Burton, A.: Understanding face familiarity. Cognition 172, 46–58 (2018)
Mohri, M., Rostamizadeh, A., Talwalkar, A.: Foundations of Machine Learning. The MIT Press, Cambridge (2012)
Park, C., Park, H.: A comparision of generalized linear discriminant analysis algorithms. Pattern Recognit. 41, 1083–1097 (2008)
Qiao, L., Chen, S., Tan, X.: Sparsity preserving projections with applications to face recognition. Pattern Recognit. 43, 331–341 (2010)
Seghouane, A., Shokouhi, N., Koch, I.: Sparse principal component analysis with preserved sparsity pattern. IEEE Trans. Image Process. 28, 3274–3285 (2019)
Shi, W., Wu, G.: Why PCA plus graph embedding methods can be unstable for extracting classification features? (submitted for publication)
Stewart, G.W.: Perturbation bounds for the definite generalized eigenvalue problem. Linear Algebra Appl. 23, 69–85 (1979)
Sun, J.: The perturbation bounds for eigenspaces of a definite matrix-pair. Numer. Math. 41, 321–343 (1983)
Tenenbaum, J., de Silva, V., Langford, J.: A global geometric framework for nonlinear dimensionality reduction. Science 290(5500), 2319–2323 (2000)
Torgerson, W.: Multidimensional scaling: I. Theory and method. Psychometrika 17, 401–419 (1952)
Turk, M., Pentland, A.: Eigenfaces for recognition. J. Cognit. Neurosci. 3, 71–86 (1991)
Van Loan, C.F.: The sensitivity of the matrix exponential. SIAM J. Numer. Anal. 14, 971–981 (1977)
Wang, J.: Geometric Structure of High-Dimensional Data and Dimensionality Reduction. Higher Education Press, Beijing (2011)
Wang, S., Chen, H., Peng, X., Zhou, C.: Exponential locality preserving projections for small sample size problem. Neurocomputing 74, 36–54 (2011)
Wang, S., Yan, S., Yang, J., et al.: A general exponential framework for dimensionality reduction. IEEE Trans. Image Process. 23, 920–930 (2014)
Webb, A.: Statistical Pattern Recognition, 2nd edn. Wiley, Hoboken (2002)
Wu, G., Feng, T., Zhang, L., Yang, M.: Inexact implementation using Krylov subspace methods for large scale exponential discriminant analysis with applications to high dimensionality reduction problems. Pattern Recogniti. 66, 328–341 (2017)
Wu, G., Xu, W., Leng, H.: Inexact and incremental bilinear Lanczos components algorithms for high dimensionality reduction and image reconstruction. Pattern Recognit. 48, 244–263 (2015)
Yan, L., Pan, J.: Two-dimensional exponential discriminant analysis and its application to face recognition. In: International Conference on Computational Aspects of Social Networks (CASoN), pp. 528–531 (2010)
Yan, S., Xu, D., Zhang, B., Zhang, H., Yang, Q., Lin, S.: Graph embedding and extensions: a general framework for dimensionality reduction. IEEE Trans. Pattern Anal. Mach. Intell. 29, 40–51 (2007)
Yang, J., Zhang, D., Yang, J., Niu, B.: Globally maximizing, locally minimizing: unsupervised discriminant projection with applications to face and palm biometrics. IEEE Trans. Pattern Anal. Mach. Intell. 29, 650–664 (2007)
Ye, J., Li, Q.: LDA/QR: an efficient and effective dimension reduction algorithm and its theoretical foundation. Pattern Recognit. 37, 851–854 (2004)
Yuan, S., Mao, X.: Exponential elastic preserving projections for facial expression recognition. Neurocomputing 275, 711–724 (2018)
Zhang, T., Fang, B., Tang, Y., Shang, Z., Xu, B.: Generalized discriminant analysis: a matrix exponential approach. IEEE Trans. Syst. Man Cybern. Part B Cybern. 40, 186–197 (2010)
Acknowledgements
We would like to express our sincere thanks to the anonymous referees and our editor for insightful comments and suggestions that greatly improved the representation of this paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work is supported by the Natural Science Foundation of Jiangsu Province under grant BK20171185, and the Fundamental Research Funds for the Central Universities of China under grant 2019XKQYMS89.
Rights and permissions
About this article
Cite this article
Shi, W., Luo, Y. & Wu, G. On general matrix exponential discriminant analysis methods for high dimensionality reduction. Calcolo 57, 18 (2020). https://doi.org/10.1007/s10092-020-00366-6
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10092-020-00366-6
Keywords
- Dimensionality reduction
- Small-sample-size problem
- Matrix exponential
- Large-scale eigenvalue problem
- Stability analysis