Abstract
Hyperbolic geometry has become popular in machine learning due to its capacity to embed discrete hierarchical graph structures with low distortions into continuous spaces for further downstream processing. It is thus becoming important to consider statistical models and inference methods for data sets grounded in hyperbolic spaces. In this work, we study the statistical f-divergences between two kinds of hyperbolic distributions: The Poincaré distributions and the related hyperboloid distributions. By exhibiting maximal invariants of group actions, we show how these f-divergences can be expressed as functions of canonical terms.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Hyperbolic geometry has constant negative curvature and the volume of hyperbolic balls increases exponentially with respect to their radii rather than polynomially as in Euclidean spaces.
References
Ali, S.M., Silvey, S.D.: A general class of coefficients of divergence of one distribution from another. J. Roy. Stat. Soc.: Ser. B (Methodol.) 28(1), 131–142 (1966)
Anderson, J.W.: Hyperbolic geometry. Springer Science & Business Media (2006)
Banerjee, A., Dhillon, I.S., Ghosh, J., Sra, S., Ridgeway, G.: Clustering on the Unit Hypersphere using von Mises-Fisher Distributions. J. Mach. Learn. Res. 6(9) (2005)
Barbaresco, F.: Lie group machine learning and gibbs density on Poincaré unit disk from souriau lie groups thermodynamics and su(1,1) coadjoint orbits. In: Nielsen, F., Barbaresco, F. (eds.) GSI 2019. LNCS, vol. 11712, pp. 157–170. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-26980-7_17
Barndorff-Nielsen, O.: Hyperbolic distributions and distributions on hyperbolae. Scand. J. Stat. 5, 151–157 (1978)
Barndorff-Nielsen, O.: Exponentially decreasing distributions for the logarithm of particle size. Proc. Royal Society of London. A. Math. Phys. Sci. 353(1674), 401–419 (1977)
Barndorff-Nielsen, O.: The hyperbolic distribution in statistical physics. Scandinavian J. Stat. 9(1), 43–46 (1982)
Barndorff-Nielsen, O.: Information and exponential families: in statistical theory. John Wiley & Sons (2014)
Barndorff-Nielsen, O., Blaesild, P.: Hyperbolic distributions and ramifications: Contributions to theory and application. In: Statistical distributions in scientific work, pp. 19–44. Springer (1981). https://doi.org/10.1007/978-94-009-8549-0_2
Blæsild, P.: The two-dimensional hyperbolic distribution and related distributions, with an application to Johannsen’s bean data. Biometrika 68(1), 251–263 (1981)
Cannon, J.W., Floyd, W.J., Kenyon, R., Parry, W.R., et al.: Hyperbolic geometry. Flavors of geometry 31(59–115), 2 (1997)
Cho, H., DeMeo, B., Peng, J., Berger, B.: Large-margin classification in hyperbolic space. In: The 22nd International Conference on Artificial Intelligence and Statistics, pp. 1832–1840. PMLR (2019)
Cohen, T., Welling, M.: Harmonic exponential families on manifolds. In: International Conference on Machine Learning, pp. 1757–1765. PMLR (2015)
Csiszár, I.: Eine informationstheoretische ungleichung und ihre anwendung auf beweis der ergodizitaet von markoffschen ketten. Magyer Tud. Akad. Mat. Kutato Int. Koezl. 8, 85–108 (1964)
Eaton, M.L.: Group invariance applications in statistics. Hayward, CA: Institute of Mathematical Statistics; Alexandria, VA: American Statistical Association (1989)
Jensen, J.L.: On the hyperboloid distribution. Scandinavian J. Stat.8(4), 193–206 (1981)
Jüttner, F.: Das Maxwellsche Gesetz der Geschwindigkeitsverteilung in der Relativtheorie. Ann. Phys. 339(5), 856–882 (1911)
Nielsen, F., Okamura, K.: Information measures and geometry of the hyperbolic exponential families of Poincaré and hyperboloid distributions. arXiv preprint arXiv:2205.13984 (2022)
Nielsen, F., Okamura, K.: On \(f\)-divergences between Cauchy distributions. IEEE Trans. Inf. Theory 69(5), 3150–3171 (2023)
Sarkar, R.: Low distortion delaunay embedding of trees in hyperbolic plane. In: van Kreveld, M., Speckmann, B. (eds.) GD 2011. LNCS, vol. 7034, pp. 355–366. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-25878-7_34
Tojo, K., Yoshino, T.: An exponential family on the upper half plane and its conjugate prior. In: Barbaresco, F., Nielsen, F. (eds.) SPIGL 2020. SPMS, vol. 361, pp. 84–95. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-77957-3_4
Tojo, K., Yoshino, T.: Harmonic exponential families on homogeneous spaces. Inform. Geometry 4(1), 215–243 (2021)
Tojo, K., Yoshino, T.: Classification problem of invariant \(q\)-exponential families on homogeneous spaces. In: Mathematical optimization and statistical theories using geometric methods OCAMI Reports, Vol. 8, pp. 85–104. Osaka Central Advanced Mathematical Institute (OCAMI) (2022). https://doi.org/10.24544/ocu.20221208-007
Verdú, S.: The Cauchy Distribution in Information Theory. entropy 25(2), 346 (2023)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Nielsen, F., Okamura, K. (2023). On the f-Divergences Between Hyperboloid and Poincaré Distributions. In: Nielsen, F., Barbaresco, F. (eds) Geometric Science of Information. GSI 2023. Lecture Notes in Computer Science, vol 14071. Springer, Cham. https://doi.org/10.1007/978-3-031-38271-0_18
Download citation
DOI: https://doi.org/10.1007/978-3-031-38271-0_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-38270-3
Online ISBN: 978-3-031-38271-0
eBook Packages: Computer ScienceComputer Science (R0)