Manifold-Aware Perturbations for Constrained Generative Modeling

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Keegan, Katherine, Ruthotto, Lars
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908800722665472
author Keegan, Katherine
Ruthotto, Lars
author_facet Keegan, Katherine
Ruthotto, Lars
contents Generative models have enjoyed widespread success in a variety of applications. However, they encounter inherent mathematical limitations in modeling distributions where samples are constrained by equalities, as is frequently the setting in scientific domains. In this work, we develop a computationally cheap, mathematically justified, and highly flexible distributional modification for combating known pitfalls in equality-constrained generative models. We propose perturbing the data distribution in a constraint-aware way such that the new distribution has support matching the ambient space dimension while still implicitly incorporating underlying manifold geometry. Through theoretical analyses and empirical evidence on several representative tasks, we illustrate that our approach consistently enables data distribution recovery and stable sampling with both diffusion models and normalizing flows.
format Preprint
id arxiv_https___arxiv_org_abs_2601_23151
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Manifold-Aware Perturbations for Constrained Generative Modeling
Keegan, Katherine
Ruthotto, Lars
Machine Learning
Generative models have enjoyed widespread success in a variety of applications. However, they encounter inherent mathematical limitations in modeling distributions where samples are constrained by equalities, as is frequently the setting in scientific domains. In this work, we develop a computationally cheap, mathematically justified, and highly flexible distributional modification for combating known pitfalls in equality-constrained generative models. We propose perturbing the data distribution in a constraint-aware way such that the new distribution has support matching the ambient space dimension while still implicitly incorporating underlying manifold geometry. Through theoretical analyses and empirical evidence on several representative tasks, we illustrate that our approach consistently enables data distribution recovery and stable sampling with both diffusion models and normalizing flows.
title Manifold-Aware Perturbations for Constrained Generative Modeling
topic Machine Learning
url https://arxiv.org/abs/2601.23151