Abstract
This chapter introduces Shannon-inspired performance limits associated with the classification of low-dimensional subspaces embedded in a high-dimensional ambient space from compressive and noisy measurements. In particular, it introduces the diversity-discrimination tradeoff that describes the interplay between the number of classes that can be separated by a compressive classifier—measured via the discrimination gain—and the performance of such a classifier—measured via the diversity gain—and the relation of such an interplay to the underlying problem geometry, including the ambient space dimension, the subspaces dimension, and the number of compressive measurements. Such a fundamental limit on performance is derived from a syntactic equivalence between the compressive classification problem and certain wireless communications problems. This equivalence provides an opportunity to cross-pollinate ideas between the wireless information theory domain and the compressive classification domain. This chapter also demonstrates how theory aligns with practice in a concrete application: face recognition from a set of noisy compressive measurements.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
We constrain the random measurement kernel to be full row rank. We also constrain the distribution of the random measurement kernel to be invariant to rotations. These constraints are obeyed by the standard Gaussian i.i.d. random kernels in compressive sensing.
- 2.
The final inequalities hold element-wise.
- 3.
- 4.
The DMT was introduced in the context of wireless communications to characterize the high-SNR performance of fading coherent MIMO channels [57, 58]. It shows that the spatial flexibility provided by multiple antennas can simultaneously increase the achievable rate and decrease the probability of error in a wireless communications channel, but only according to a tradeoff that is tightly characterized at high SNR.
- 5.
By working with the upper bound to the misclassification probability rather than the true one, we obtain a lower bound to the diversity gain rather than the exact one.
- 6.
Note that this classifier is mismatched in view of the fact that the class conditioned distributions are not necessarily Gaussian. In fact, it is immediate to demonstrate that face samples within each class do not pass the Royston’s multivariate normality test [41], as they return p-values below 10−3 for all classes.
- 7.
Note that this suggests that nature may tend to approximately distribute the subspaces uniformly on the Grassmann manifold.
References
Adini, Y., Moses, Y., Ullman, S.: Face recognition: the problem of compensating for changes in illumination direction. IEEE Trans. Pattern Anal. Mach. Intell. 19(7), 721–732 (1997)
Aggarwal, V., Ashikhmin, A., Calderbank, R.: A Grassmannian packing based on the Nordstrom-Robinson code. In: IEEE Information Theory Workshop, Chengdu, China, pp. 1–5 (2006)
Alon, U., Barkai, N., Notterman, D.A., Gish, K., Ybarra, S., Mack, D., Levine, A.J.: Broad patterns of gene expression revealed by clustering analysis of tumor and normal colon tissues probed by oligonucleotide arrays. Proc. Natl. Acad. Sci. 96(12), 6745–6750 (1999)
Ashikhmin, A., Calderbank, R.: Space-time Reed-Muller codes for noncoherent MIMO transmission. In: IEEE International Symposium on Information Theory, Adelaide, Australia, 1952–1956 (2005)
Ashikhmin, A., Calderbank, R., Kewlin, W.: Multidimensional second order Reed-Muller codes as Grassmannian packings. In: IEEE International Symposium on Information Theory, Seattle, WA, USA, pp. 1001–1005 (2006)
Ashikhmin, A., Calderbank, R.: Grassmannian packings from operator Reed-Muller codes. IEEE Trans. Inf. Theory 56(10), 5689–5714 (2010)
Basri, R., Jacobs, D.W.: Lambertian reflectance and linear subspaces. IEEE Trans. Pattern Anal. Mach. Intell. 25(2), 218–233 (2003)
Calderbank, R., Hardin, R.H., Rains, E.M., Shor, P.W., Sloane, N.J.A.: A group-theoretic framework for the construction of packings in Grassmannian spaces. J. Algebraic Comb. 9, 129–140 (1999)
Candès, E., Romberg, J., Tao, T.: Robust uncertainty principles: Exact signal reconstruction from highly incomplete frequency information. IEEE Trans. Inf. Theory 52(2), 489–509 (2006)
Candès, E., Romberg, J., Tao, T.: Stable signal recovery from incomplete and inaccurate measurements. Commun. Pure Appl. Math. 59(8), 1207–1223, 2006.
Candès, E., Tao, T.: Near-optimal signal recovery from random projections: universal encoding strategies? IEEE Trans. Inf. Theory 52(12), 5406–5425 (2006)
Carson, W.R., Chen, M., Rodrigues, M.R.D., Calderbank, R., Carin, L.: Communications-inspired projection design with application to compressive sensing. SIAM J. Imaging Sci. 5(4), 1185–1212 (2012)
Chen, M., Silva, J., Paisley, J., Wang, C., Dunson, D., Carin, L.: Compressive sensing on manifolds using a nonparametric mixture of factor analyzers: algorithm and performance bounds. IEEE Trans. Signal Process. 58(12), 6140–6155 (2010)
Chen, M., Carson, W.R., Rodrigues, M.R.D., Calderbank, R., Carin, L.: Communication-inspired linear discriminant analysis. In: International Conference on Machine Learning, Edinburgh, UK, pp. 919–926 (2012)
Davenport, M., Boufounos, P., Wakin, M., Baraniuk, R.: Signal processing with compressive measurements. IEEE J. Sel. Top. Sign. Process. 4(2), 445–460 (2010)
Donoho, D.: Compressed sensing. IEEE Trans. Inf. Theory 52(4), 1289–1306 (2006)
Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification, 2nd edn. Wiley-Interscience, New York, NY (2000)
Elhamifar, Vidal, R.: Sparse subspace clustering: algorithm, theory, and applications. IEEE Trans. Pattern Anal. Mach. Intell. 35(11), 2765–2781 (2013)
Georghiades, A.S., Belhumeur, P.N., Kriegman, D.J.: From few to many: Illumination cone models for face recognition under variable lighting and pose. IEEE Trans. Pattern Anal. Mach. Intell. 23(6), 643–660 (2001)
Hastie, T., Simard, P.Y.: Metrics and models for handwritten character recognition. Stat. Sci. 13(1), 54–65 (1998)
Hild, K., Erdogmus, D., Torkkola, K., Principe, J.: Feature extraction using information-theoretic learning. IEEE Trans. Pattern Anal. Mach. Intell. 28(9), 1385–1392 (2006)
Hochwald, B.M., Marzetta, T.L.: Unitary space-time modulation for multiple-antenna communications in Rayleigh flat fading. IEEE Trans. Inf. Theory 46(2), 543–564 (2000)
Hochwald, B.M., Marzetta, T.L., Richardson, T.J., Sweldens, W., Urbanke, R.: Systematic design of unitary spce-time constellations. IEEE Trans. Inf. Theory 46(6), 1962–1973 (2000)
Hull, J.J.: A database for handwritten text recognition research. IEEE Trans. Pattern Anal. Mach. Intell. 16(5), 550–554 (1994)
Kaski, S., Peltonen, J.: Informative discriminant analysis. In: International Conference on Machine Learning, Washington, DC, USA, pp. 329–336 (2003)
LeCun, Y., Jackel, L., Bottou, L., Brunot, A., Cortes, C., Denker, J., Drucker, H., Guyon, I., Muller, U., Sackinger, E., Simard, P., Vapnik, V.: Comparison of learning algorithms for handwritten digit recognition. In: International Conference on Artificial Neural Networks, Warsaw, Poland, pp. 53–60 (1995)
Lee, K.-C., Ho, J., Kriegman, D.J.: Acquiring linear subspaces for face recognition under variable lighting. IEEE Trans. Pattern Anal. Mach. Intell. 27(5), 684–698 (2005)
Liu, G., Lin, Z., Yu, Y.: Robust subspace segmentation by low-rank representation. In: International Conference on Machine Learning, Haifa, Israel, pp. 663–670 (2010)
Liu, L., Fieguth, P.: Texture classification from random features. IEEE Trans. Pattern Anal. Mach. Intell. 34(3), 574–586 (2012)
Marzetta, T.L., Hochwald, B.M.: Capacity of a mobile multiple-antenna communication link in Rayleigh flat fading. IEEE Trans. Inf. Theory 45(1), 139–157 (1999)
Nenadic, Z.: Information discriminant analysis: feature extraction with an information-theoretic objective. IEEE Trans. Pattern Anal. Mach. Intell. 29(8), 1394–1407 (2007)
Nokleby, M., Rodrigues, M.R.D., Calderbank, R.: Discrimination on the Grassmann manifold: fundamental limits of subspace classifiers. Available at http://arxiv.org/abs/1404.5187
Qiu, Q., Sapiro, G.: Learning robust subspace clustering. Available at http://arxiv.org/abs/1308.0273
Qiu, Q., Sapiro, G.: Learning transformations for clustering and classification. Available at http://arxiv.org/abs/1309.2074
Qiu, Q., Sapiro, G.: Learning transformations for classification forests. Available at http://arxiv.org/abs/1312.5604
Reboredo, H., Renna, F., Calderbank, R., Rodrigues, M.R.D.: Compressive classification of a mixture of Gaussians: analysis, designs and geometrical interpretation. Available at http://arxiv.org/abs/1401.6962
Reeves, G., Gastpar, M.: The sampling rate distortion tradeoff for sparsity pattern recovery in compressed sensing. IEEE Trans. Inf. Theory 58(5), 3065–3092 (2012)
Reeves, G., Gastpar, M.: Approximate sparsity pattern recovery: information-theoretic lower bounds. IEEE Trans. Inf. Theory 59(6), 3451–3465 (2013)
Renna, F., Calderbank, R., Carin, L., Rodrigues, M.R.D.: Reconstruction of signals drawn from a Gaussian mixture from noisy compressive measurements. IEEE Trans. Signal Process. 62(9), 2265–2277 (2014)
Ross, D.T., et al.: Systematic variation in gene expression patterns in human cancer cell lines. Nat. Genet. 24(3), 227–235 (2000)
Royston, J.P.: Some techniques for assessing multivariate normality based on the Shapiro-Wilk W. Appl. Stat. 32(2), 121–133 (1983)
Soltanolkotabi, M., Candès, E.: A geometric analysis of subspace clustering with outliers. Ann. Stat. 40(4), 2195–2238 (2012)
Soltanolkotabi, M., Elhamifar, E., Candès, E.: Robust subspace clustering. Available at arxiv.org/abs/1301.2603
Tao, D., Li, X., Wu, X., Maybank, S.: Geometric mean for subspace selection. IEEE Trans. Pattern Anal. Mach. Intell. 31(2), 260–274 (2009)
Torkkola, K.: Learning discriminative feature transforms to low dimensions in low dimensions. In: Advances in Neural Information Processing Systems, Vancouver, Canada, pp. 969–976 (2001)
Torkkola, K.: Feature extraction by non-parametric mutual information maximization. J. Mach. Learn. Res. 3, 1415–1438 (2003)
Tenenbaum, J., de Silva, V., Langford, J.: A global geometric framework for nonlinear dimensionality reduction. Science 290(5500), 2319–2323 (2000)
Tulino, A., Caire, G., Verdú, S., Shamai, S.: Support recovery with sparsely sampled free random matrices. IEEE Trans. Inf. Theory 59(7), 4243–4271 (2013)
Wainwright, M.: Sharp thresholds for high-dimensional and noisy sparsity recovery using l1-constrained quadratic programming (lasso). IEEE Trans. Inf. Theory 55(5), 2183–2202 (2009)
Wainwright, M.: Information-theoretic limits on sparsity recovery in the high dimensional and noisy setting. IEEE Trans. Inf. Theory 55(12), 5728–5741 (2009)
Wang, W., Wainwright, M., Ramchandran, K.: Information-theoretic limits on sparse signal recovery: Dense versus sparse measurement matrices. IEEE Trans. Inf. Theory 56(6), 2967–2979 (2010)
Wang, L., Carlson, D., Rodrigues, M.R.D., Wilcox, D., Calderbank, R., Carin, L.: Designed measurements for vector count data. In: Advances in Neural Information Processing Systems, Lake Tahoe, NV, USA, pp. 1142–1150 (2013)
Wang, L., Razi, A., Rodrigues, M.R.D., Calderbank, R., Carin, L.: Nonlinear information-theoretic compressive measurement design. In: International Conference on Machine Learning, Beijing, China, pp. 1161–1169 (2014)
Wang, Y., Xu, H.: Noisy sparse subspace clustering. In: International Conference on Machine Learning, Atlanta, GA, USA, pp. 89–97 (2013)
Wright, J., Yang, M., Ganesh, A., Sastry, S., Ma, Y.: Robust face recognition via sparse representation. IEEE Trans. Pattern Anal. Mach. Intell. 31(2), 210–227 (2009)
Zheng, L., Tse, D.: The diversity-multiplexing tradeoff for non-coherent multiple antenna channels. In: The Annual Allerton Conference on Communication, Control and Computing, Monticello, IL, USA, pp. 1011–1020 (2002)
Zheng, L., Tse, D.: Communication on the Grassmann manifold: a geometric approach to the noncoherent multiple-antenna channel. IEEE Trans. Inf. Theory 48(2), 359–383 (2002)
Zheng, L., Tse, D.: Diversity and multiplexing: a fundamental tradeoff in multiple-antenna channels. IEEE Trans. Inf. Theory 49(5), 1073–1096 (2003)
Acknowledgements
This work was supported by the Royal Society International Exchanges Scheme IE120996. The work of Robert Calderbank and Matthew Nokleby is also supported in part by the Air Force Office of Scientific Research under the Complex Networks Program.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Rodrigues, M., Nokleby, M., Renna, F., Calderbank, R. (2015). Compressive Classification: Where Wireless Communications Meets Machine Learning. In: Boche, H., Calderbank, R., Kutyniok, G., Vybíral, J. (eds) Compressed Sensing and its Applications. Applied and Numerical Harmonic Analysis. Birkhäuser, Cham. https://doi.org/10.1007/978-3-319-16042-9_15
Download citation
DOI: https://doi.org/10.1007/978-3-319-16042-9_15
Publisher Name: Birkhäuser, Cham
Print ISBN: 978-3-319-16041-2
Online ISBN: 978-3-319-16042-9
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)