Abstract
In this paper, first, we survey the concept of diffeological Fisher metric and its naturality, using functorial language of probabilistic morphisms, and slightly extending Lê’s theory in [Le2020] to include weakly \(C^k\)-diffeological statistical models. Then we introduce the resulting notions of the diffeological Fisher distance, the diffeological Hausdorff–Jeffrey measure and explain their role in classical and Bayesian nonparametric estimation problems in statistics.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
In this paper we shall consider only (possibly weakly) Fréchet differentiable mappings and we shall omit “Fréchet” in the remaining part of this paper.
References
Ay, N., Jost, J., Lê, H.V., Schwachhöfer, L.: Information geometry and sufficient statistics. Probab. Theory Relat. Fields 162, 327–364 (2015)
Ay, N., Jost, J., Lê, H.V., Schwachhöfer, L.: Information Geometry. Springer, Cham (2017)
Ay, N., Jost, J., Lê, H.V., Schwachhöfer, L.: Parametrized measure models. Bernoulli 24, 1692–1725 (2018)
Amari, S.: The Finsler geometry of families nonregular distributions. Translated into English by S. V. Sabau in 2002 (1984)
Amari, S.: Information Geometry and Its Applications. Applied Mathematical Sciences, vol. 194. Springer, Berlin (2016)
Ambrisio, L., Tilli, P.: Topics in Analysis on Metric Spaces. Oxford University Press, Oxford (2004)
Barbaresco, F., Gay-Balmaz, F.: Lie group cohomology and (multi)symplectic integrators: new geometric tools for Lie group machine learning based on Souriau geometric statistical mechanics. Entropy 22, 498 (2020). https://doi.org/10.3390/e22050498
Bogachev, V.I.: Differentiable Measures and the Malliavin Calculus. AMS, Providence (2010)
Bogachev, V.I.: Weak convergence of measures, Mathematical Surveys and Monographs, vol. 234. American Mathematical Society, Providence (2018)
Borovkov, A.A.: Mathematical Statistics. Gordon and Breach Science Publishers, Amsterdam (1998)
Faden, A.M.: The existence of regular conditional probabilities: necessary and sufficient conditions. Ann. Probab. 13, 288–298 (1985)
Federer, H.: Geometric Measure Theory, Die Grundlehren der mathematischen Wissenschaften, vol. 153. Springer, New York (1969)
Friedrich, T.: Die Fisher-Information und symplektische Strukturen. Math. Nachr. 153, 273–296 (1991)
Giry, M.: A categorical approach to probability theory. In: Banaschewski, B. (ed.) Categorical Aspects of Topology and Analysis. Lecture Notes in Mathematics, vol. 915, pp. 68–85. Springer, Heidelberg (1982)
Chentsov, N.: Statistical Decision Rules and Optimal Inference. Nauka, Moscow, Russia (1972). English translation in: Translation of Math. Monograph, vol. 53. American Mathematical Society, Providence, RI, USA, 1982
Iglesias-Zemmour, P.: Diffeology. American Mathematical Society, Providence (2013)
Jeffrey, H.: An invariant form for the prior probability in estimation problems. Proc. Ro. Soc. Lond. Ser. Math. Phys. Sci. 186, 453–461 (1946)
Jost, J., Lê, H.V., Tran, T.D.: Probabilistic morphisms and Bayesian nonparametrics. Eur. Phys. J. Plus 136, 441 (2021). https://doi.org/10.1140/epjp/s13360-021-01427-7
Jordan, M.I.: What are the open problems in Bayesian statistics? ISBA Bull. 18, 1–4 (2011)
Kaliaj, S.B.: Differentiability and weak differentiability. Mediterr. J. Math. 13, 2801–2811 (2016). https://doi.org/10.1007/s00009-015-0656-6
Lawvere, W.F.: The category of probabilistic mappings (1962). Unpublished. https://ncatlab.org/nlab/files/lawvereprobability1962.pdf
Lê, H.V.: The uniqueness of the Fisher metric as information metric. Ann. Inst. Stat. Math. 69, 879–896 (2017). arXiv:math/1306.1465
Lê, H.V.: Diffeological statistical models, the Fisher metric and probabilistic mappings. Mathematics 8(2), 167 (2020). https://doi.org/10.3390/math8020167
Morgan, F.: Geometric Measure Theory A Beginner’s Guide, 4th edn. Elsevier, Amsterdam (2009)
McCullagh, P.: What is a statistical model. Ann. Stat. 30, 1225–1310 (2002)
Morse, N., Sacksteder, R.: Statistical isomorphism. Ann. Math. Stat. 37, 203–214 (1966)
Neveu, J.: Bases Mathématiques du Calcul de Probabilités, deuxième edition, Masson, Paris (1970)
Pflug, G.: Optimization of Stochastic Models. Kluwer Academic, Boston (1996)
Pflug, G.: Derivatives of probability measures - concepts and applications to the optimization of stochastic systems. In: Lecture Notes in Control and Information Science, vol. 103, pp. 252–274. Springer, Heidelberg (1988)
Souriau, J.-M.: Groupes différentiels. In: Lecture Notes in Mathematics, vol. 836, pp. 91–128. Springer, Heidelberg (1980)
Watanabe, S.: Algebraic Geometry and Statistical Learning Theory. Cambridge University Press, Cambridge (2009)
Acknowledgement
HVL would like to thank Professor Frédéric Barbaresco and Professor Frank Nielsen for their invitation to the Les Houches conference in July 2020 where a part of the results in this paper has been reported. She is grateful to Professor Sun-ichi Amari who discussed with her the phenomena of degeneration and explosion of the Fisher metric and sent her a copy of his paper [Amari1984] several years ago. The authors would like to thank the referee for helpful comments which considerably improve the exposition of the present paper.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Lê, H.V., Tuzhilin, A.A. (2021). Nonparametric Estimations and the Diffeological Fisher Metric. In: Barbaresco, F., Nielsen, F. (eds) Geometric Structures of Statistical Physics, Information Geometry, and Learning. SPIGL 2020. Springer Proceedings in Mathematics & Statistics, vol 361. Springer, Cham. https://doi.org/10.1007/978-3-030-77957-3_7
Download citation
DOI: https://doi.org/10.1007/978-3-030-77957-3_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-77956-6
Online ISBN: 978-3-030-77957-3
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)