A probabilistic view on Riemannian machine learning models for SPD matrices

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: de Surrel, Thibault, Yger, Florian, Lotte, Fabien, Chevallier, Sylvain
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917056092307456
author de Surrel, Thibault
Yger, Florian
Lotte, Fabien
Chevallier, Sylvain
author_facet de Surrel, Thibault
Yger, Florian
Lotte, Fabien
Chevallier, Sylvain
contents The goal of this paper is to show how different machine learning tools on the Riemannian manifold $\mathcal{P}_d$ of Symmetric Positive Definite (SPD) matrices can be united under a probabilistic framework. For this, we will need several Gaussian distributions defined on $\mathcal{P}_d$. We will show how popular classifiers on $\mathcal{P}_d$ can be reinterpreted as Bayes Classifiers using these Gaussian distributions. These distributions will also be used for outlier detection and dimension reduction. By showing that those distributions are pervasive in the tools used on $\mathcal{P}_d$, we allow for other machine learning tools to be extended to $\mathcal{P}_d$.
format Preprint
id arxiv_https___arxiv_org_abs_2505_02402
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A probabilistic view on Riemannian machine learning models for SPD matrices
de Surrel, Thibault
Yger, Florian
Lotte, Fabien
Chevallier, Sylvain
Machine Learning
Statistics Theory
The goal of this paper is to show how different machine learning tools on the Riemannian manifold $\mathcal{P}_d$ of Symmetric Positive Definite (SPD) matrices can be united under a probabilistic framework. For this, we will need several Gaussian distributions defined on $\mathcal{P}_d$. We will show how popular classifiers on $\mathcal{P}_d$ can be reinterpreted as Bayes Classifiers using these Gaussian distributions. These distributions will also be used for outlier detection and dimension reduction. By showing that those distributions are pervasive in the tools used on $\mathcal{P}_d$, we allow for other machine learning tools to be extended to $\mathcal{P}_d$.
title A probabilistic view on Riemannian machine learning models for SPD matrices
topic Machine Learning
Statistics Theory
url https://arxiv.org/abs/2505.02402