Salvato in:
Dettagli Bibliografici
Autori principali: Fré, Pietro Giuseppe, Milanesio, Federico, Sanguinetti, Guido, Santoro, Matteo
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2507.16871
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913955241263104
author Fré, Pietro Giuseppe
Milanesio, Federico
Sanguinetti, Guido
Santoro, Matteo
author_facet Fré, Pietro Giuseppe
Milanesio, Federico
Sanguinetti, Guido
Santoro, Matteo
contents Recent work has identified non-compact symmetric spaces U/H as a promising class of homogeneous manifolds to develop a geometrically consistent theory of neural networks. An initial implementation of these concepts has been presented in a twin paper under the moniker of Cartan Neural Networks, showing both the feasibility and the performance of these geometric concepts in a machine learning context. The current paper expands on the mathematical structures underpinning Cartan Neural Networks, detailing the geometric properties of the layers and how the maps between layers interact with such structures to make Cartan Neural Networks covariant and geometrically interpretable. Together, these twin papers constitute a first step towards a fully geometrically interpretable theory of neural networks exploiting group-theoretic structures
format Preprint
id arxiv_https___arxiv_org_abs_2507_16871
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Navigation through Non-Compact Symmetric Spaces: a mathematical perspective on Cartan Neural Networks
Fré, Pietro Giuseppe
Milanesio, Federico
Sanguinetti, Guido
Santoro, Matteo
Machine Learning
High Energy Physics - Theory
Recent work has identified non-compact symmetric spaces U/H as a promising class of homogeneous manifolds to develop a geometrically consistent theory of neural networks. An initial implementation of these concepts has been presented in a twin paper under the moniker of Cartan Neural Networks, showing both the feasibility and the performance of these geometric concepts in a machine learning context. The current paper expands on the mathematical structures underpinning Cartan Neural Networks, detailing the geometric properties of the layers and how the maps between layers interact with such structures to make Cartan Neural Networks covariant and geometrically interpretable. Together, these twin papers constitute a first step towards a fully geometrically interpretable theory of neural networks exploiting group-theoretic structures
title Navigation through Non-Compact Symmetric Spaces: a mathematical perspective on Cartan Neural Networks
topic Machine Learning
High Energy Physics - Theory
url https://arxiv.org/abs/2507.16871