Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Ziheng, Wu, Xiao-Jun, Schölkopf, Bernhard, Sebe, Nicu
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2509.07115
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912594763186176
author Chen, Ziheng
Wu, Xiao-Jun
Schölkopf, Bernhard
Sebe, Nicu
author_facet Chen, Ziheng
Wu, Xiao-Jun
Schölkopf, Bernhard
Sebe, Nicu
contents Normalization layers are crucial for deep learning, but their Euclidean formulations are inadequate for data on manifolds. On the other hand, many Riemannian manifolds in machine learning admit gyro-structures, enabling principled extensions of Euclidean neural networks to non-Euclidean domains. Inspired by this, we introduce GyroBN, a principled Riemannian batch normalization framework for gyrogroups. We establish two necessary conditions, namely \emph{pseudo-reduction} and \emph{gyroisometric gyrations}, that guarantee GyroBN with theoretical control over sample statistics, and show that these conditions hold for all known gyrogroups in machine learning. Our framework also incorporates several existing Riemannian normalization methods as special cases. We further instantiate GyroBN on seven representative geometries, including the Grassmannian, five constant curvature spaces, and the correlation manifold, and derive novel gyro and Riemannian structures to enable these instantiations. Experiments across these geometries demonstrate the effectiveness of GyroBN. The code is available at https://github.com/GitZH-Chen/GyroBN.git.
format Preprint
id arxiv_https___arxiv_org_abs_2509_07115
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Riemannian Batch Normalization: A Gyro Approach
Chen, Ziheng
Wu, Xiao-Jun
Schölkopf, Bernhard
Sebe, Nicu
Machine Learning
Artificial Intelligence
Normalization layers are crucial for deep learning, but their Euclidean formulations are inadequate for data on manifolds. On the other hand, many Riemannian manifolds in machine learning admit gyro-structures, enabling principled extensions of Euclidean neural networks to non-Euclidean domains. Inspired by this, we introduce GyroBN, a principled Riemannian batch normalization framework for gyrogroups. We establish two necessary conditions, namely \emph{pseudo-reduction} and \emph{gyroisometric gyrations}, that guarantee GyroBN with theoretical control over sample statistics, and show that these conditions hold for all known gyrogroups in machine learning. Our framework also incorporates several existing Riemannian normalization methods as special cases. We further instantiate GyroBN on seven representative geometries, including the Grassmannian, five constant curvature spaces, and the correlation manifold, and derive novel gyro and Riemannian structures to enable these instantiations. Experiments across these geometries demonstrate the effectiveness of GyroBN. The code is available at https://github.com/GitZH-Chen/GyroBN.git.
title Riemannian Batch Normalization: A Gyro Approach
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2509.07115