Mechanics of geodesics in Information geometry

12/14/2022
by   Sumanto Chanda, et al.
0

In this article we attempt to formulate Riemannian and Randers-Finsler metrics in information geometry and study their mechanical properties. Starting from the gradient flow equations, we show how to formulate Riemannian metrics, and demonstrate their duality under canonical transformation. Then we show how to formulate a Randers-Finsler metric from deformed gradient equations. The theories described are finally applied to the Gaussian model and tested to verify consistency.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/30/2022

Weyl geometry and the gradient-flow equations in information geometry

The gradient-flow equations in information geometry is reconsidered from...
research
05/20/2020

Riemannian geometry for Compound Gaussian distributions: application to recursive change detection

A new Riemannian geometry for the Compound Gaussian distribution is prop...
research
11/17/2017

Dimensionality Reduction on Grassmannian via Riemannian Optimization: A Generalized Perspective

This paper proposes a generalized framework with joint normalization whi...
research
04/02/2020

Mirrorless Mirror Descent: A More Natural Discretization of Riemannian Gradient Flow

We present a direct (primal only) derivation of Mirror Descent as a "par...
research
04/02/2021

Hybrid and Generalized Bayesian Cramér-Rao Inequalities via Information Geometry

Information geometry is the study of statistical models from a Riemannia...
research
03/09/2023

Scalable Stochastic Gradient Riemannian Langevin Dynamics in Non-Diagonal Metrics

Stochastic-gradient sampling methods are often used to perform Bayesian ...
research
09/21/2021

Geometric Fabrics: Generalizing Classical Mechanics to Capture the Physics of Behavior

Classical mechanical systems are central to controller design in energy ...

Please sign up or login with your details

Forgot password? Click here to reset