Fisher-Rao geometry of Dirichlet distributions

05/12/2020
by   Alice Le Brigant, et al.
0

In this paper, we study the geometry induced by the Fisher-Rao metric on the parameter space of Dirichlet distributions. We show that this space is geodesically complete and has everywhere negative sectional curvature. An important consequence of this negative curvature for applications is that the Fréchet mean of a set of Dirichlet distributions is uniquely defined in this geometry.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/17/2019

The Fisher-Rao geometry of beta distributions applied to the study of canonical moments

This paper studies the Fisher-Rao geometry on the parameter space of bet...
research
06/10/2022

Fisher SAM: Information Geometry and Sharpness Aware Minimisation

Recent sharpness-aware minimisation (SAM) is known to find flat minima w...
research
02/02/2018

A reversal phenomenon in estimation based on multiple samples from the Poisson--Dirichlet distribution

Consider two forms of sampling from a population: (i) drawing s samples ...
research
07/06/2020

On the minmax regret for statistical manifolds: the role of curvature

Model complexity plays an essential role in its selection, namely, by ch...
research
06/03/2020

Classifying histograms of medical data using information geometry of beta distributions

In this paper, we use tools of information geometry to compare, average ...
research
05/27/2019

Probabilistic mappings and Bayesian nonparametrics

In this paper we develop a functorial language of probabilistic mappings...
research
03/06/2023

The Matrix-variate Dirichlet Averages and Its Applications

This paper is about Dirichlet averages in the matrix-variate case or ave...

Please sign up or login with your details

Forgot password? Click here to reset