Guardado en:
Detalles Bibliográficos
Autores principales: Elhag, Ahmed A., Wang, Yuyang, Susskind, Joshua M., Bautista, Miguel Angel
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:https://arxiv.org/abs/2305.15586
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866907771025227776
author Elhag, Ahmed A.
Wang, Yuyang
Susskind, Joshua M.
Bautista, Miguel Angel
author_facet Elhag, Ahmed A.
Wang, Yuyang
Susskind, Joshua M.
Bautista, Miguel Angel
contents We present Manifold Diffusion Fields (MDF), an approach that unlocks learning of diffusion models of data in general non-Euclidean geometries. Leveraging insights from spectral geometry analysis, we define an intrinsic coordinate system on the manifold via the eigen-functions of the Laplace-Beltrami Operator. MDF represents functions using an explicit parametrization formed by a set of multiple input-output pairs. Our approach allows to sample continuous functions on manifolds and is invariant with respect to rigid and isometric transformations of the manifold. In addition, we show that MDF generalizes to the case where the training set contains functions on different manifolds. Empirical results on multiple datasets and manifolds including challenging scientific problems like weather prediction or molecular conformation show that MDF can capture distributions of such functions with better diversity and fidelity than previous approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2305_15586
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Manifold Diffusion Fields
Elhag, Ahmed A.
Wang, Yuyang
Susskind, Joshua M.
Bautista, Miguel Angel
Machine Learning
We present Manifold Diffusion Fields (MDF), an approach that unlocks learning of diffusion models of data in general non-Euclidean geometries. Leveraging insights from spectral geometry analysis, we define an intrinsic coordinate system on the manifold via the eigen-functions of the Laplace-Beltrami Operator. MDF represents functions using an explicit parametrization formed by a set of multiple input-output pairs. Our approach allows to sample continuous functions on manifolds and is invariant with respect to rigid and isometric transformations of the manifold. In addition, we show that MDF generalizes to the case where the training set contains functions on different manifolds. Empirical results on multiple datasets and manifolds including challenging scientific problems like weather prediction or molecular conformation show that MDF can capture distributions of such functions with better diversity and fidelity than previous approaches.
title Manifold Diffusion Fields
topic Machine Learning
url https://arxiv.org/abs/2305.15586