Advertisement

Supervised Deep Canonical Correlation Analysis for Multiview Feature Learning

  • Yan Liu
  • Yun LiEmail author
  • Yun-Hao YuanEmail author
  • Ji-Peng Qiang
  • Min Ruan
  • Zhao Zhang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10639)

Abstract

Recently, a new feature representation method called deep canonical correlation analysis (DCCA) has been proposed with high learning performance for multiview feature extraction of high dimensional data. DCCA is an effective approach to learn the nonlinear mappings of two sets of random variables that make the resulting DNN representations highly correlated. However, the DCCA learning process is unsupervised and thus lacks the class label information of training samples on the two views. In order to take full advantage of the class information of training samples, we propose a discriminative version of DCCA referred to as supervised DCCA (SDCCA) for feature learning, which explicitly considers the class information of samples. Compared with DCCA, the SDCCA method can not only guarantee the nonlinear maximal correlation between two views, but also minimize within-class scatter of the samples. With supervision, SDCCA can extract more discriminative features for pattern classification tasks. We test SDCCA on the handwriting recognition and speech recognition using two popular MNIST and XRMB datasets. Experimental results show that SDCCA gets higher performance than several related algorithms.

Keywords

Deep learning Canonical correlations Multiview learning 

Notes

Acknowledgements

This work is supported by National Natural Science Foundation of China under Grant No. 61402203. In addition, it is also supported in part by the National Natural Science Foundation of China under Grant Nos. 61472344, 61611540347, the Natural Science Foundation of Jiangsu Province of China under Grant Nos. BK20161338, BK20170513, and sponsored by Excellent Young Backbone Teacher Project.

References

  1. 1.
    Hotelling, H.: Relations between two sets of variates. Biometrika 28(3/4), 321–377 (1936)CrossRefzbMATHGoogle Scholar
  2. 2.
    Sun, Q.S., Liu, Z.-D., Heng, P.-A., Xia, D.-S.: A theorem on the generalized canonical projective vectors. Pattern Recogn. 38(3), 449–452 (2005)CrossRefzbMATHGoogle Scholar
  3. 3.
    Lai, P.L., Fyfe, C.: Kernel and nonlinear canonical correlation analysis. Int. J. Neural Syst. 10(5), 365 (2000)CrossRefGoogle Scholar
  4. 4.
    Andrew, G., Arora, R., Bilmes, J., Livescu, K.: Deep canonical correlation analysis. In: ICML, pp. III-1247–III-1255. JMLR: W&CP, Atlanta (2013)Google Scholar
  5. 5.
    Yan, F., Mikolajczyk, K.: Deep correlation for matching images and text. In: CVPR, pp. 3441–3451. IEEE, Boston (2015)Google Scholar
  6. 6.
    Wang, W., Yan, X., Lee, H., Livescu, K.: Deep variational canonical correlation analysis arXiv preprint arXiv:1610.03454v3 (2017)
  7. 7.
    Wang, W., Arora, R., Livescu, K., Bilmes, J.: On deep multi-view representation learning. In: ICML, pp. 1083–1092. JMLR: W&CP, Lille (2015)Google Scholar
  8. 8.
    Melzer, T., Reiter, M., Bischof, H.: Appearance models based on kernel canonical correlation analysis. Pattern Recogn. 36(9), 1961–1971 (2003)CrossRefzbMATHGoogle Scholar
  9. 9.
    Vincent, P., Larochelle, H., Lajoie, I., Bengio, Y., Manzagol, P.-A.: Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion. J. Mach. Learn. Res. 11, 3371–3408 (2010)zbMATHMathSciNetGoogle Scholar
  10. 10.
    Michaeli, T., Wang, W., Livescu, K.: Nonparametric canonical correlation analysis. In: ICML, pp. 1967–1976. JMLR: W&CP, New York (2016)Google Scholar
  11. 11.
    Chang, C.-C., Lin, C.-J.: LIBSVM: a library for support vector machines. ACM TIST 2(3) (2011). Article No. 27Google Scholar
  12. 12.
    Wang, W., Arora, R., Livescu, K., Bilmes, J.: Unsupervised learning of acoustic features via deep canonical correlation analysis. In: ICASSP, pp. 4590–4594. IEEE, Brisbane (2015)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.School of Information EngineeringYangzhou UniversityYangzhouChina

Personalised recommendations