Abstract
The Fisher-Rao distance between two probability distribution functions, as well as other divergence measures, is related to entropy and is in the core of the research area of information geometry. It can provide a framework and enlarge the perspective of analysis for a wide variety of domains, such as statistical inference, image processing (texture classification and inpainting), clustering processes and morphological classification. We present here a compact summary of results regarding the Fisher-Rao distance in the space of multivariate normal distributions including some historical background, closed forms in special cases, bounds, numerical approaches and references to recent applications.
Supported by FAPESP (13/25997-7) and CNPq (313326/2017-7) foundations.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Costa, S.I.R., Santos, S.A., Strapasson, J.E.: Fisher information distance: a geometrical reading. Discret. Appl. Math. 197, 59–69 (2015)
Mahalanobis, P.C.: On the generalized distance in statistics. Proc. Natl. Inst. Sci. 2, 49–55 (1936)
Bhattacharyya, A.: On a measure of divergence between two statistical populations defined by their probability distributions. Bull. Calcutta Math. Soc. 35, 99–110 (1943)
Hotelling, H.: Spaces of statistical parameters. Bull. Am. Math. Soc. (AMS) 36, 191 (1930)
Rao, C.R.: Information and the accuracy attainable in the estimation of statistical parameters. Bull. Calcutta Math. Soc. 37, 81–91 (1945)
Fisher, R.A.: On the mathematical foundations of theoretical statistics. Philos. Trans. R. Soc. Lond. 222, 309–368 (1921)
Burbea, J.: Informative geometry of probability spaces. Expositiones Mathematica 4, 347–378 (1986)
Atkinson, C., Mitchell, A.F.S.: Rao’s distance measure. Samkhyã Indian J. Stat. 43, 345–365 (1981)
Angulo, J., Velasco-Forero, S.: Morphological processing of univariate Gaussian distribution-valued images based on Poincaré upper-half plane representation. In: Nielsen, F. (ed.) Geometric Theory of Information, pp. 331–366. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-05317-2_12
Maybank, S.J., Ieng, S., Benosman, R.: A Fisher-Rao metric for paracatadioptric images of lines. Int. J. Comput. Vis. 99(2), 147–165 (2012)
Schwander, O., Nielsen, F.: Model centroids for the simplification of kernel density estimators. In: IEEE - Acoustics, Speech and Signal Processing (ICASSP) (2012)
Taylor, S.: Clustering financial return distributions using the Fisher information metric. Entropy 21(2), 110 (2019)
Skovgaard, L.T.: A Riemannian geometry of the multivariate normal model. Scand. J. Stat. 11, 211–223 (1984)
Efron, B.: Defining the curvature of a statistical problem (with applications to second order efficiency). Ann. Stat. 3, 1189–1242 (1975)
Dawid, A.P.: Discussions to Efron’s paper. Ann. Stat. 3, 1231–1234 (1975)
Amari, S., Nagaoka, H.: Differential Geometrical Methods in Statistics. Lecture Notes in Statistics, vol. 28. Springer, New York Heidelberg (1986). https://doi.org/10.1007/978-1-4612-5056-2
Amari, S., Nagaoka, H.: Methods of Information Geometry. Translations of Mathematical Monographs, vol. 191. American Mathematical Society, New York (2000)
Chentsov, N.N.: Statistical Decision Rules and Optimal Inference, vol. 53. AMS Bookstore, New York (1982)
Nielsen, F.: An elementary introduction to information geometry. arXiv preprint arXiv:1808.08271 (2018)
Amari, S.: Information Geometry and Its Applications, vol. 194. Springer, Tokyo (2016). https://doi.org/10.1007/978-4-431-55978-8
Strapasson, J.E., Porto, J., Costa, S.I.R.: On bounds for the Fisher-Rao distance between multivariate normal distributions. In: Bayesian Inference and Maximum Entropy Methods in Science and Engineering (MAXENT 2014). AIP, vol. 1641 (2015)
Strapasson, J.E., Pinele, J., Costa, S.I.R.: A totally geodesic submanifold of the multivariate normal distributions and bounds for the Fisher-Rao distance. In: IEEE Information Theory Workshop (ITW) (2016)
Strapasson, J.E., Pinele, J., Costa, S.I.R.: Clustering using the Fisher-Rao distance. In: Sensor Array and Multichannel Signal Processing Workshop. IEEE (2016)
Eriksen, P.S.: Geodesics connected with the fischer metric on the multivariate normal manifold. Aalborg University Centre, Institute of Electronic Systems (1986)
Calvo, M., Oller, J.M.: An explicit solution of information geodesic equations for the multivariate normal model. Stat. Decis. 9, 119–138 (1991)
Han, M., Park, F.C.: DTI segmentation and fiber tracking using metrics on multivariate normal distributions. J. Math. Imaging Vis. 49(2), 317–334 (2014)
Lenglet, C., Rousson, M., Deriche, R., Faugeras, O.: Statistics on the manifold of multivariate normal distributions: theory and application to diffusion tensor MRI processing. J. Math. Imaging Vis. 25(3), 423–444 (2006)
Moakher, M., Mourad, Z.: The Riemannian geometry of the space of positive-definite matrices and its application to the regularization of positive-definite matrix-valued data. J. Math. Imaging Vis. 40(2), 171–187 (2011)
Verdoolaege, G., Scheunders, P.: Geodesics on the manifold of multivariate generalized Gaussian distributions with an application to multicomponent texture discrimination. Int. J. Comput. Vis. 95(3), 265 (2011)
Gattone, S., et al.: On the geodesic distance in shapes K-means clustering. Entropy 20(9), 647 (2018)
Garcia, V., Nielsen, F.: Simplification and hierarchical representations of mixtures of exponential families. Signal Process. 90(12), 3197–3212 (2010)
Nielsen, F., The statistical Minkowski distances: closed-form formula for Gaussian mixture models. arXiv preprint arXiv:1901.03732 (2019)
Calvo, M., Oller, J.M.: A distance between multivariate normal distributions based in an embedding into the Siegel group. J. Multivar. Anal. 35(2), 223–242 (1990)
Pilté, M., Barbaresco, F.: Tracking quality monitoring based on information geometry and geodesic shooting. In: Radar Symposium (IRS). IEEE (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Pinele, J., Costa, S.I.R., Strapasson, J.E. (2019). On the Fisher-Rao Information Metric in the Space of Normal Distributions. In: Nielsen, F., Barbaresco, F. (eds) Geometric Science of Information. GSI 2019. Lecture Notes in Computer Science(), vol 11712. Springer, Cham. https://doi.org/10.1007/978-3-030-26980-7_70
Download citation
DOI: https://doi.org/10.1007/978-3-030-26980-7_70
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-26979-1
Online ISBN: 978-3-030-26980-7
eBook Packages: Computer ScienceComputer Science (R0)