Horoball Hulls and Extents in Positive Definite Space

  • P. Thomas Fletcher
  • John Moeller
  • Jeff M. Phillips
  • Suresh Venkatasubramanian
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6844)

Abstract

The space of positive definite matrices P(n) is a Riemannian manifold with variable nonpositive curvature. It includes Euclidean space and hyperbolic space as submanifolds, and poses significant challenges for the design of algorithms for data analysis. In this paper, we develop foundational geometric structures and algorithms for analyzing collections of such matrices. A key technical contribution of this work is the use of horoballs, a natural generalization of halfspaces for non-positively curved Riemannian manifolds. We propose generalizations of the notion of a convex hull and a centerpoint and approximations of these structures using horoballs and based on novel decompositions of P(n). This leads to an algorithm for approximate hulls using a generalization of extents.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Agarwal, P.K., Har-Peled, S., Varadarajan, K.R.: Approximating extent measures of points. JACM 51 (2004)Google Scholar
  2. 2.
    Arnaudon, M., Nielsen, F.: On Approximating the Riemannian 1-Center, arXiv:1101.4718v1 (2011)Google Scholar
  3. 3.
    Basser, P.J., Mattiello, J., LeBihan, D.: MR diffusion tensor spectroscopy and imaging. Biophys. J. 66(1), 259–267 (1994)CrossRefGoogle Scholar
  4. 4.
    Belkin, M., Niyogi, P.: Semi-supervised learning on Riemannian manifolds. Machine Learning 56(1), 209–239 (2004)CrossRefMATHGoogle Scholar
  5. 5.
    Bhatia, R.: Positive Definite Matrices. Princeton University Press, Princeton (2006)MATHGoogle Scholar
  6. 6.
    Billera, L., Holmes, S., Vogtmann, K.: Geometry of the Space of Phylogenetic Trees. Advances in Applied Mathematics 27(4), 733–767 (2001)MathSciNetCrossRefMATHGoogle Scholar
  7. 7.
    Bonnabel, S., Sepulchre, R.: Riemannian metric and geometric mean for positive semidefinite matrices of fixed rank. SIAM J. on Matr. Anal. & App. 31(3), 1055–1070 (2010)MathSciNetCrossRefMATHGoogle Scholar
  8. 8.
    Bridson, M.R., Haefliger, A.: Metric Spaces of Non-Positive Curvature. Springer, Heidelberg (2009)MATHGoogle Scholar
  9. 9.
    Chepoi, V., Dragan, F., Estellon, B., Habib, M., Vaxès, Y.: Diameters, centers, and approximating trees of delta-hyperbolicgeodesic spaces and graphs. In: SoCG (2008)Google Scholar
  10. 10.
    Chepoi, V., Estellon, B.: Packing and covering δ-hyperbolic spaces by balls. In: Charikar, M., Jansen, K., Reingold, O., Rolim, J.D.P. (eds.) RANDOM 2007 and APPROX 2007. LNCS, vol. 4627, pp. 59–73. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  11. 11.
    Cowin, S.: The structure of the linear anisotropic elastic symmetries. Journal of the Mechanics and Physics of Solids 40, 1459–1471 (1992)MathSciNetCrossRefMATHGoogle Scholar
  12. 12.
    Dudley, R.M.: Metric entropy of some classes of sets with differentiable boundaries. Journal of Approximation Theory 10, 227–236 (1974)MathSciNetCrossRefMATHGoogle Scholar
  13. 13.
    Dyer, R., Zhang, H., Möller, T.: Surface sampling and the intrinsic Voronoi diagram. In: SGP, pp. 1393–1402 (2008)Google Scholar
  14. 14.
    Eppstein, D.: Squarepants in a tree: Sum of subtree clustering and hyperbolic pants decomposition. ACM Transactions on Algorithms (TALG) 5 (2009)Google Scholar
  15. 15.
    Fletcher, P.T., Joshi, S.: Principal geodesic analysis on symmetric spaces: Statistics of diffusion tensors. Com. Vis. & Math. Methods Med. Biomed. Im. Anal, 87–98 (2004)Google Scholar
  16. 16.
    Fletcher, P.T., Venkatasubramanian, S., Joshi, S.: The geometric median on Riemannian manifolds with application to robust atlas estimation. NeuroImage 45, S143–S152 (2009)CrossRefGoogle Scholar
  17. 17.
    Gromov, M.: Hyperbolic groups. Essays in Group Theory 8, 75–263 (1987)MathSciNetCrossRefMATHGoogle Scholar
  18. 18.
    Krauthgamer, R., Lee, J.R.: Algorithms on negatively curved spaces. In: FOCS (2006)Google Scholar
  19. 19.
    Leibon, G., Letscher, D.: Delaunay triangulations and Voronoi diagrams for Riemannian manifolds. In: SoCG (2000)Google Scholar
  20. 20.
    Meyer, G., Bonnabel, S., Sepulchre, R.: Regression on fixed-rank positive semidefinite matrices: a Riemannian approach, arXiv:1006.1288 (2010)Google Scholar
  21. 21.
    Moakher, M.: A Differential Geometric Approach to the Geometric Mean of Symmetric Positive-Definite Matrices. SIAM J. on Matr. Anal. & App. 26 (2005)Google Scholar
  22. 22.
    Moakher, M., Batchelor, P.G.: Symmetric positive-definite matrices: From geometry to applications and visualization. In: Visualization and Processing of Tensor Fields (2006)Google Scholar
  23. 23.
    Nesterov, Y., Todd, M.: On the Riemannian geometry defined by self-concordant barriers and interior-point methods. FoCM 2(4), 333–361 (2008)MathSciNetMATHGoogle Scholar
  24. 24.
    Owen, M., Provan, J.: A Fast Algorithm for Computing Geodesic Distances in Tree Space, arXiv:0907.3942 (2009)Google Scholar
  25. 25.
    Pennec, X., Fillard, P., Ayache, N.: A Riemannian framework for tensor computing. IJCV 66(1), 41–66 (2006)CrossRefMATHGoogle Scholar
  26. 26.
    Shawe-Taylor, J., Cristianini, N.: Kernel Methods for Pattern Analysis, Cambridge (2004)Google Scholar
  27. 27.
    Smith, S.: Covariance, subspace, and intrinsic Cramér-Rao bounds. IEEE Transactions on Signal Processing 53(5), 1610–1630 (2005)MathSciNetCrossRefGoogle Scholar
  28. 28.
    Vandereycken, B., Absil, P., Vandewalle, S.: A Riemannian geometry with complete geodesics for the set of positive semidefinite matrices of fixed rank. status: submitted (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • P. Thomas Fletcher
    • 1
  • John Moeller
    • 1
  • Jeff M. Phillips
    • 1
  • Suresh Venkatasubramanian
    • 1
  1. 1.University of UtahUnited States

Personalised recommendations