Abstract
Problems involving high-dimensional data, such as pattern recognition, image analysis, and gene clustering, often require a preliminary step of dimension reduction before or during statistical analysis. If one restricts to a linear technique for dimension reduction, the remaining issue is the choice of the projection. This choice can be dictated by desire to maximize certain statistical criteria, including variance, kurtosis, sparseness, and entropy, of the projected data. Motivations for such criteria comes from past empirical studies of statistics of natural and urban images. We present a geometric framework for finding projections that are optimal for obtaining certain desired statistical properties. Our approach is to define an objective function on spaces of orthogonal linear projections—Stiefel and Grassmann manifolds, and to use gradient techniques to optimize that function. This construction uses the geometries of these manifolds to perform the optimization. Experimental results are presented to demonstrate these ideas for natural and facial images.
Similar content being viewed by others
References
Bell, A.J., Sejnowski, T.J.: An information maximization approach to blind separation and blind deconvolution. Neural Comput. 7, 1129–1159 (1995)
Comon, P.: Independent component analysis, a new concept? Signal Process. Special issue on higher-order statistics 36(3), 287–314 (1994)
Cook, D.: Testing predictor contributions in sufficient dimension reduction. Ann. Stat. 32(3), 1062–1092 (2004)
Cook, D., Li, B.: Dimension reduction for conditional mean in regression. Ann. Stat. 30(2), 455–474 (2002)
Donoho, D.L., Flesia, A.G.: Can recent innovations in harmonic analysis “explain” key findings in natural image statistics? Netw. Comput. Neural Syst. 12(3), 371–393 (2001)
Edelman, A., Arias, T.A., Smith, S.T.: The geometry of algorithms with orthogonality constraints. SIAM J. Matrix Anal. Appl. 20(2), 303–353 (1998)
Field, D.J.: What is the goal of sensory coding? Neural Comput. 6(4), 559–601 (1994)
Fiori, S.: A minor subspace algorithm based on neural Stiefel dynamics. Int. J. Neural Syst. 19(5), 339–350 (2002a)
Fiori, S.: A theory for learning based on rigid bodies dynamics. IEEE Trans. Neural Netw. 13(3), 521–531 (2002b)
Geman, S., Hwang, C.-R.: Diffusions for global optimization. SIAM J. Control Optim. 24(5), 1031–1043 (1987)
Golub, G.H., Van Loan, C.: Matrix Computations. The John Hopkins University Press, Baltimore (1989)
Hyvärinen, A.: Fast and robust fixed-point algorithm for independent component analysis. IEEE Trans. Neural Netw. 10(3), 626–634 (1999)
Hyvärinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. Wiley, New York (2001)
Johnson, R.A., Wichern, D.W.: Applied Multivariate Statistical Analysis. Prentice Hall, New York (2001)
Liu, X., Srivastava, A., Gallivan, K.A.: Optimal linear representations of images for object recognition. In: Proceedings of 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 229–234 (2003)
Mallat, S.G.: Theory for multiresolution signal decomposition: The wavelet representation. IEEE Trans. Pattern Anal. Mach. Intell. 11(7), 674–693 (1989)
Olshausen, B.A., Field, D.J.: Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381, 607–609 (1996a)
Olshausen, B.A., Field, D.J.: Natural image statistics and efficient coding. Netw. Comput. Neural Syst. 7, 333–339 (1996b)
Robert, C.P., Casella, G.: Monte Carlo Statistical Methods. Springer Texts in Statistics. Springer, New York (1999)
Srivastava, A.: A Bayesian approach to geometric subspace estimation. IEEE Trans. Signal Process. 48(5), 1390–1400 (2000)
Srivastava, A., Lee, A.B., Simoncelli, E.P., Zhu, S.-C.: On advances in statistical modeling of natural images. J. Math. Imaging Vis. 18, 17–33 (2003)
Srivastava, A., Liu, X.: Tools for application-driven linear dimension reduction. J. Neurocomput. 67, 136–160 (2005)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Rubinshtein, E., Srivastava, A. Optimal linear projections for enhancing desired data statistics. Stat Comput 20, 267–282 (2010). https://doi.org/10.1007/s11222-009-9120-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-009-9120-4