Hessian Structures and Divergence Functions on Deformed Exponential Families

Chapter

Abstract

A Hessian structure \((\nabla , h)\) on a manifold is a pair of a flat affine connection \(\nabla \) and a semi-Riemannian metric \(h\) which is given by a Hessian of some function. In information geometry, it is known that an exponential family naturally has dualistic Hessian structures and their canonical divergences coincide with the Kullback-Leibler divergences, which are also called the relative entropies. A deformed exponential family is a generalization of exponential families. A deformed exponential family naturally has two kinds of dualistic Hessian structures and conformal structures of Hessian metrics. In this paper, geometry of such Hessian structures and conformal structures are summarized. In addition, divergence functions on these Hessian manifolds are constructed from the viewpoint of estimating functions. As an application of such Hessian structures to statistics, a generalization of independence and geometry of generalized maximum likelihood method are studied.

Keywords

Hessian manifold Statistical manifold Deformed exponential family Divergence  Information geometry Tsallis statistics 

References

  1. 1.
    Amari, S., Nagaoka, H.: Method of Information Geometry. American Mathematical Society, Providence, Oxford University Press, Oxford (2000)Google Scholar
  2. 2.
    Amari, S., Ohara, A.: Geometry of q-exponential family of probability distributions. Entropy 13, 1170–1185 (2011)CrossRefMathSciNetGoogle Scholar
  3. 3.
    Amari, S., Ohara, A., Matsuzoe, H.: Geometry of deformed exponential families: invariant, dually-flat and conformal geometry. Phys. A. 391, 4308–4319 (2012)CrossRefMathSciNetGoogle Scholar
  4. 4.
    Barondorff-Nielsen, O.E., Jupp, P.E.: Statistics, yokes and symplectic geometry. Ann. Facul. Sci. Toulouse 6, 389–427 (1997)CrossRefGoogle Scholar
  5. 5.
    Basu, A., Harris, I.R., Hjort, N.L., Jones, M.C.: Robust and efficient estimation by minimising a density power divergence. Biometrika 85, 549–559 (1998)CrossRefMATHMathSciNetGoogle Scholar
  6. 6.
    Borgesa, E.P.: A possible deformed algebra and calculus inspired in nonextensive thermostatistics. Phys. A 340, 95–101 (2004)CrossRefMathSciNetGoogle Scholar
  7. 7.
    Eguchi, S.: Geometry of minimum contrast. Hiroshima Math. J. 22, 631–647 (1992)MATHMathSciNetGoogle Scholar
  8. 8.
    Fujimoto, Y., Murata, N.: A generalization of independence in naive bayes model. Lect. Notes Comp. Sci. 6283, 153–161 (2010)CrossRefGoogle Scholar
  9. 9.
    Fujimoto Y., Murata N.: A generalisation of independence in statistical models for categorical distribution. Int. J. Data Min. Model. Manage. 2(4), 172–187 (2012)Google Scholar
  10. 10.
    Ivanov, S.: On dual-projectively flat affine connections. J. Geom. 53, 89–99 (1995)CrossRefMATHMathSciNetGoogle Scholar
  11. 11.
    Kumon, M., Takemura, A., Takeuchi, K.: Conformal geometry of statistical manifold with application to sequential estimation. Sequential Anal. 30, 308–337 (2011)CrossRefMATHMathSciNetGoogle Scholar
  12. 12.
    Kurose, T.: On the divergences of 1-conformally flat statistical manifolds. Tôhoku Math. J. 46, 427–433 (1994)CrossRefMATHMathSciNetGoogle Scholar
  13. 13.
    Kurose, T.: Conformal-projective geometry of statistical manifolds. Interdiscip. Inform. Sci. 8, 89–100 (2002)MATHMathSciNetGoogle Scholar
  14. 14.
    Lauritzen, S. L.: Statistical Manifolds, Differential Geometry in Statistical Inferences, IMS Lecture Notes Monograph Series, vol. 10, pp. 96–163. Hayward, California (1987)Google Scholar
  15. 15.
    Matsuzoe, H.: Geometry of contrast functions and conformal geometry. Hiroshima Math. J. 29, 175–191 (1999)MATHMathSciNetGoogle Scholar
  16. 16.
    Matsuzoe, H.: Geometry of statistical manifolds and its generalization. In: Proceedings of the 8th International Workshop on Complex Structures and Vector Fields, pp. 244–251. World Scientific, Singapore (2007)Google Scholar
  17. 17.
    Matsuzoe, H.: Computational geometry from the viewpoint of affine differential geometry. Lect. Notes Comp. Sci. 5416, 103–113 (2009)CrossRefGoogle Scholar
  18. 18.
    Matsuzoe, H.: Statistical manifolds and geometry of estimating functions, pp. 187–202. Recent Progress in Differential Geometry and Its Related Fields World Scientific, Singapore (2013)Google Scholar
  19. 19.
    Matsuzoe, H., Henmi, M.: Hessian structures on deformed exponential families. Lect. Notes Comp. Sci. 8085, 275–282 (2013)CrossRefGoogle Scholar
  20. 20.
    Matsuzoe, H., Ohara, A.: Geometry for \(q\)-exponential families. In: Recent progress in differential geometry and its related fields, pp. 55–71. World Scientific, Singapore (2011)Google Scholar
  21. 21.
    Murata, N., Takenouchi, T., Kanamori, T., Eguchi, S.: Information geometry of u-boost and bregman divergence. Neural Comput. 16, 1437–1481 (2004)CrossRefMATHGoogle Scholar
  22. 22.
    Naudts, J.: Estimators, escort probabilities, and \(\phi \)-exponential families in statistical physics. J. Ineq. Pure Appl. Math. 5, 102 (2004)Google Scholar
  23. 23.
    Naudts, J.: Generalised Thermostatistics, Springer, New York (2011)Google Scholar
  24. 24.
    Ohara, A.: Geometric study for the legendre duality of generalized entropies and its application to the porous medium equation. Euro. Phys. J. B. 70, 15–28 (2009)CrossRefMATHGoogle Scholar
  25. 25.
    Ohara, A., Matsuzoe H., Amari S.: Conformal geometry of escort probability and its applications. Mod. Phys. Lett. B. 10, 26:1250063 (2012)Google Scholar
  26. 26.
    Ohara A., Wada, T.: Information geometry of q-Gaussian densities and behaviors of solutions to related diffusion equations. J. Phys. A: Math. Theor. 43, 035002 (2010)Google Scholar
  27. 27.
    Okamoto, I., Amari, S., Takeuchi, K.: Asymptotic theory of sequential estimation procedures for curved exponential families. Ann. Stat. 19, 961–961 (1991)CrossRefMATHMathSciNetGoogle Scholar
  28. 28.
    Shima, H.: The Geometry of Hessian Structures, World Scientific, Singapore (2007)Google Scholar
  29. 29.
    Suyari, H., Tsukada, M.: Law of error in tsallis statistics. IEEE Trans. Inform. Theory 51, 753–757 (2005)CrossRefMathSciNetGoogle Scholar
  30. 30.
    Takatsu, A.: Behaviors of \(\varphi \)-exponential distributions in wasserstein geometry and an evolution equation. SIAM J. Math. Anal. 45, 2546–2546 (2013)CrossRefMATHMathSciNetGoogle Scholar
  31. 31.
    Tanaka, M.: Meaning of an escort distribution and \(\tau \)-transformation. J. Phys.: Conf. Ser. 201, 012007 (2010)Google Scholar
  32. 32.
    Tsallis, C.: Possible generalization of boltzmann—gibbs statistics. J. Stat. Phys. 52, 479–487 (1988)CrossRefMATHMathSciNetGoogle Scholar
  33. 33.
    Tsallis, C.: Introduction to Nonextensive Statistical Mechanics: Approaching a Complex World. Springer, New York (2009)Google Scholar
  34. 34.
    Vigelis, R.F., Cavalcante, C.C.: On \(\phi \)-families of probability distributions. J. Theor. Probab. 21, 1–25 (2011)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  1. 1.Department of Computer Science and Engineering, Graduate School of EngineeringNagoya Institute of TechnologyNagoyaJapan
  2. 2.The Institute of Statistical MathematicsTachikawaJapan

Personalised recommendations