Synergizing Transport-Based Generative Models and Latent Geometry for Stochastic Closure Modeling

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Dong, Xinghao, Yang, Huchen, Wu, Jin-long
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917281582284800
author Dong, Xinghao
Yang, Huchen
Wu, Jin-long
author_facet Dong, Xinghao
Yang, Huchen
Wu, Jin-long
contents Diffusion models recently developed for generative AI tasks can produce high-quality samples while still maintaining diversity among samples to promote mode coverage, providing a promising path for learning stochastic closure models. Compared to other types of generative AI models, such as GANs and VAEs, the sampling speed is known as a key disadvantage of diffusion models. By systematically comparing transport-based generative models on a numerical example of 2D Kolmogorov flows, we show that flow matching in a lower-dimensional latent space is suited for fast sampling of stochastic closure models, enabling single-step sampling that is up to two orders of magnitude faster than iterative diffusion-based approaches. To control the latent space distortion and thus ensure the physical fidelity of the sampled closure term, we compare the implicit regularization offered by a joint training scheme against two explicit regularizers: metric-preserving (MP) and geometry-aware (GA) constraints. Besides offering a faster sampling speed, both explicitly and implicitly regularized latent spaces inherit the key topological information from the lower-dimensional manifold of the original complex dynamical system, which enables the learning of stochastic closure models without demanding a huge amount of training data.
format Preprint
id arxiv_https___arxiv_org_abs_2602_17089
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Synergizing Transport-Based Generative Models and Latent Geometry for Stochastic Closure Modeling
Dong, Xinghao
Yang, Huchen
Wu, Jin-long
Machine Learning
Dynamical Systems
Computational Physics
Diffusion models recently developed for generative AI tasks can produce high-quality samples while still maintaining diversity among samples to promote mode coverage, providing a promising path for learning stochastic closure models. Compared to other types of generative AI models, such as GANs and VAEs, the sampling speed is known as a key disadvantage of diffusion models. By systematically comparing transport-based generative models on a numerical example of 2D Kolmogorov flows, we show that flow matching in a lower-dimensional latent space is suited for fast sampling of stochastic closure models, enabling single-step sampling that is up to two orders of magnitude faster than iterative diffusion-based approaches. To control the latent space distortion and thus ensure the physical fidelity of the sampled closure term, we compare the implicit regularization offered by a joint training scheme against two explicit regularizers: metric-preserving (MP) and geometry-aware (GA) constraints. Besides offering a faster sampling speed, both explicitly and implicitly regularized latent spaces inherit the key topological information from the lower-dimensional manifold of the original complex dynamical system, which enables the learning of stochastic closure models without demanding a huge amount of training data.
title Synergizing Transport-Based Generative Models and Latent Geometry for Stochastic Closure Modeling
topic Machine Learning
Dynamical Systems
Computational Physics
url https://arxiv.org/abs/2602.17089