Abstract
This paper proposes an efficient supervised Matrix Entropy Driven Maximum Margin Feature Learning method (M3FL) to optimize all the discriminative features simultaneously. Specifically, we first present an in-depth investigation to the heteroscedastic problem in the maximum margin criterion, and then propose a new Maximum Margin Framework (MMF) based on the analysis to improve the traditional maximum margin criterion. The proposed MMF is robust to the initialization by exploring the \(\ell _1\)-norm property. We further analyze the proposed MMF and find that it is necessary to learn the projection matrix from the perspective of matrix entropy. Consequently, the M3FL method is proposed to make the matrix entropy of the projection matrix as small as possible, and the corresponding optimization algorithm is developed. In addition, we discuss the relationship between the proposed optimization algorithm w.r.t. M3FL and the optimization algorithm w.r.t. MMF. Experiments are conducted on six widely-used data sets and experimental results demonstrate that the proposed method outperforms the state-of-the-art methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bhattacharjee, A., et al.: Classification of human lung carcinomas by mRNA expression profiling reveals distinct adenocarcinoma subclasses. Proc. Nat. Acad. Sci. USA 98(24), 13790–13795 (2001)
Fodor, S.P.A.: DNA sequencing: massively parallel genomics. Science 277(5324), 393–395 (1997)
Hou, C., Jiao, Y., Nie, F., Luo, T., Zhou, Z.: 2D feature selection by sparse matrix regression. IEEE Trans. Image Process. 26(9), 4255–4268 (2017)
Li, B.N., Yu, Q., Wang, R., Xiang, K., Wang, M., Li, X.: Block principal component analysis with nongreedy l1-norm maximization. IEEE Trans. Cybern. 46(11), 2543–2547 (2016)
Li, B., Zheng, C.H., Huang, D.S.: Locally linear discriminant embedding: an efficient method for face recognition. Pattern Recogn. 41(12), 3813–3821 (2008)
Li, C., Liu, Q., Dong, W., Wei, F., Zhang, X., Yang, L.: Max-margin-based discriminative feature learning. IEEE Trans. Neural Netw. 27(12), 2768–2775 (2016)
Li, C., Shao, Y., Deng, N.Y.: Robust l1-norm two-dimensional linear discriminant analysis. Neural Netw. 65(C), 92–104 (2015)
Li, H., Jiang, T., Zhang, K.: Efficient and robust feature extraction by maximum margin criterion. IEEE Trans. Neural Netw. 17(1), 157–165 (2006)
Li, M., Yuan, B.: 2D-LDA: a statistical linear discriminant analysis for image matrix. Pattern Recogn. Lett. 26(5), 527–532 (2005)
Li, Z., Tang, J.: Unsupervised feature selection via nonnegative spectral analysis and redundancy control. IEEE Trans. Image Process. 24(12), 5343–5355 (2015)
Li, Z., Tang, J., He, X.: Robust structured nonnegative matrix factorization for image representation. IEEE Trans. Neural Netw. 29, 1947–1960 (2018)
Liu, C., Wechsler, H.: Gabor feature based classification using the enhanced fisher linear discriminant model for face recognition. IEEE Trans. Image Process. 11(4), 467–476 (2002)
Liu, J., Chen, S., Tan, X., Zhang, D.: Comments on efficient and robust feature extraction by maximum margin criterion. IEEE Trans. Neural Netw. 18(6), 1862–1864 (2007)
Liu, Y., Gao, Q., Miao, S., Gao, X., Nie, F., Li, Y.: A non-greedy algorithm for l1-norm LDA. IEEE Trans. Image Process. 26(2), 684–695 (2016)
Lu, G.F., Tang, G., Zou, J.: Spare l1-norm-based maximum margin criterion. J. Vis. Commun. Image Represent. 38(C), 11–17 (2016)
Muller, H., Stadtmuller, U.: Estimation of heteroscedasticity in regression analysis. Ann. Stat. 15(2), 610–625 (1987)
Nie, F., Yuan, J., Huang, H.: Optimal mean robust principal component analysis. In: International Conference on Machine Learning, pp. 1062–1070 (2014)
Price, A.L., Patterson, N.J., Plenge, R.M., Weinblatt, M.E., Shadick, N.A., Reich, D.: Principal components analysis corrects for stratification in genome-wide association studies. Nat. Genet. 38(8), 904–907 (2006)
Si, C., Dao, C., Bin, L.: L1-norm-based maximum margin criterion. Chin. J. Electron. 44(6), 1383–1388 (2016)
Singh, D., et al.: Gene expression correlates of clinical prostate cancer behavior. Cancer Cell 1(2), 203–209 (2002)
Tang, J., Li, Z., Wang, M., Zhao, R.: Neighborhood discriminant hashing for large-scale image retrieval. IEEE Trans. Image Process. 24(9), 2827–2840 (2015)
Tang, J., et al.: Tri-clustered tensor completion for social-aware image tag refinement. IEEE Trans. Pattern Anal. Mach. Intell. 39(8), 1662–1674 (2017)
Tao, D., Li, X., Wu, X., Maybank, S.J.: General tensor discriminant analysis and gabor features for gait recognition. IEEE Trans. Pattern Anal. Mach. Intell. 29(10), 1700–1715 (2007)
Wan, M., Lai, Z.: Multi-manifold locality graph embedding based on the maximum margin criterion (MLGE/MMC) for face recognition. IEEE Access 5, 9823–9830 (2017)
Wang, H., Lu, X., Hu, Z., Zheng, W.: Fisher discriminant analysis with l1-norm. IEEE Trans. Cybern. 44(6), 828–842 (2013)
Wang, Q., Gao, Q.: Two-dimensional PCA with f-norm minimization. In: Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence, pp. 2718–2724 (2017)
Xie, J., Jian, Y., Qian, J., Ying, T., Zhang, H.: Robust nuclear norm-based matrix regression with applications to robust face recognition. IEEE Trans. Image Process. 5(99), 2286–2295 (2017)
Xu, D., Yan, S., Tao, D., Lin, S., Zhang, H.: Marginal fisher analysis and its variants for human gait recognition and content-based image retrieval. IEEE Trans. Image Process. 16(11), 2811–2821 (2007)
Yang, K.J., Cai, Z., Li, J., Lin, G.: A stable gene selection in microarray data analysis. BMC Bioinform. 7(1), 228–244 (2006)
Ye, Q., Yang, J., Liu, F., Zhao, C., Ye, N., Yin, T.: L1-norm distance linear discriminant analysis based on an effective iterative algorithm. IEEE Trans. Circuits Syst. Video Technol. 28(1), 114–129 (2018)
Ye, Q., Yang, J., Yin, T., Zhang, Z.: Can the virtual labels obtained by traditional LP approaches be well encoded in WLR. IEEE Trans. Neural Netw. 27(7), 1591–1598 (2016)
Ye, Q., Yin, T., Gao, S., Jing, J., Zhang, Y., Sun, C.: Recursive dimension reduction for semisupervised learning. Neurocomputing 171, 1629–1636 (2016)
Zhang, D., Zhang, L., Ye, Q., Ruan, H.: Robust learning-based prediction for timber-volume of living trees. Comput. Electron. Agric. 136, 97–110 (2017)
Zhang, Z., Chow, T.W.S.: Robust linearly optimized discriminant analysis. Neurocomputing 79, 140–157 (2012)
Zhang, Z., Chow, W.S.: Tensor locally linear discriminative analysis. IEEE Sig. Process. Lett. 18(11), 643–646 (2011)
Zhang, Z., Yan, S., Zhao, M.: Pairwise sparsity preserving embedding for unsupervised subspace learning and classification. IEEE Trans. Image Process. 22(12), 4640–4651 (2013)
Zheng, W., Lai, J., Li, S.Z.: 1D-LDA vs. 2D-LDA: when is vector-based linear discriminant analysis better than matrix-based? Pattern Recogn. 41(7), 2156–2172 (2008)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Zhang, D., Tang, J., Li, Z. (2018). Matrix Entropy Driven Maximum Margin Feature Learning. In: Geng, X., Kang, BH. (eds) PRICAI 2018: Trends in Artificial Intelligence. PRICAI 2018. Lecture Notes in Computer Science(), vol 11012. Springer, Cham. https://doi.org/10.1007/978-3-319-97304-3_29
Download citation
DOI: https://doi.org/10.1007/978-3-319-97304-3_29
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-97303-6
Online ISBN: 978-3-319-97304-3
eBook Packages: Computer ScienceComputer Science (R0)