Buchumschlag

Transferable foundation models for geometric tasks on point cloud representations: geometric neural operators

Gespeichert in:
Veröffentlicht in:Machine Learning : Science and Technology vol. 6, no. 4 (Dec 2025), p. 045045
1. Verfasser: Quackenbush, B
Weitere Verfasser: Atzberger, P J
Veröffentlicht:
IOP Publishing
Schlagworte:
Data processing
Partial differential equations
Curvature
Machine learning
Differential geometry
Operators (mathematics)
Numerical methods
Cognitive tasks
Representations
Topology
Online-Zugang:Citation/Abstract
Full Text - PDF
Tags: Tag hinzufügen
Keine Tags, Fügen Sie das erste Tag hinzu!
Abstract:We introduce methods for obtaining pretrained geometric neural operators (GNPs) that can serve as basal foundation models for use in obtaining geometric features. These can be used within data processing pipelines for machine learning tasks and numerical methods. We show how our GNPs can be trained to learn robust latent representations for the differential geometry of point-clouds to provide estimates of metric, curvature, and other shape-related features. We demonstrate how our pre-trained GNPs can be used (i) to estimate the geometric properties of surfaces of arbitrary shape and topologies with robustness in the presence of noise, (ii) to approximate solutions of geometric partial differential equations on manifolds, and (iii) to solve equations for shape deformations such as curvature driven flows. We release codes and weights for using GNPs in the package <ext-link ext-link-type="uri" xlink3ahref="https://github.com/atzberg/geo_neural_op">geo_neural_op</ext-link>. This allows for incorporating our pre-trained GNPs as components for reuse within existing and new data processing pipelines. The GNPs also can be used as part of numerical solvers involving geometry or as part of methods for performing inference and other geometric tasks.
ISSN:2632-2153
DOI:10.1088/2632-2153/ae1bf8
Quelle:Science Database