AdS-GNN -- a Conformally Equivariant Graph Neural Network

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhdanov, Maksim, Iqbal, Nabil, Bekkers, Erik, Forré, Patrick
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912381994532864
author Zhdanov, Maksim
Iqbal, Nabil
Bekkers, Erik
Forré, Patrick
author_facet Zhdanov, Maksim
Iqbal, Nabil
Bekkers, Erik
Forré, Patrick
contents Conformal symmetries, i.e.\ coordinate transformations that preserve angles, play a key role in many fields, including physics, mathematics, computer vision and (geometric) machine learning. Here we build a neural network that is equivariant under general conformal transformations. To achieve this, we lift data from flat Euclidean space to Anti de Sitter (AdS) space. This allows us to exploit a known correspondence between conformal transformations of flat space and isometric transformations on the AdS space. We then build upon the fact that such isometric transformations have been extensively studied on general geometries in the geometric deep learning literature. We employ message-passing layers conditioned on the proper distance, yielding a computationally efficient framework. We validate our model on tasks from computer vision and statistical physics, demonstrating strong performance, improved generalization capacities, and the ability to extract conformal data such as scaling dimensions from the trained network.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12880
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AdS-GNN -- a Conformally Equivariant Graph Neural Network
Zhdanov, Maksim
Iqbal, Nabil
Bekkers, Erik
Forré, Patrick
Machine Learning
Artificial Intelligence
High Energy Physics - Theory
Conformal symmetries, i.e.\ coordinate transformations that preserve angles, play a key role in many fields, including physics, mathematics, computer vision and (geometric) machine learning. Here we build a neural network that is equivariant under general conformal transformations. To achieve this, we lift data from flat Euclidean space to Anti de Sitter (AdS) space. This allows us to exploit a known correspondence between conformal transformations of flat space and isometric transformations on the AdS space. We then build upon the fact that such isometric transformations have been extensively studied on general geometries in the geometric deep learning literature. We employ message-passing layers conditioned on the proper distance, yielding a computationally efficient framework. We validate our model on tasks from computer vision and statistical physics, demonstrating strong performance, improved generalization capacities, and the ability to extract conformal data such as scaling dimensions from the trained network.
title AdS-GNN -- a Conformally Equivariant Graph Neural Network
topic Machine Learning
Artificial Intelligence
High Energy Physics - Theory
url https://arxiv.org/abs/2505.12880