Vectorized Conditional Neural Fields: A Framework for Solving Time-dependent Parametric Partial Differential Equations

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Hagnberger, Jan, Kalimuthu, Marimuthu, Musekamp, Daniel, Niepert, Mathias
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913429747400704
author Hagnberger, Jan
Kalimuthu, Marimuthu
Musekamp, Daniel
Niepert, Mathias
author_facet Hagnberger, Jan
Kalimuthu, Marimuthu
Musekamp, Daniel
Niepert, Mathias
contents Transformer models are increasingly used for solving Partial Differential Equations (PDEs). Several adaptations have been proposed, all of which suffer from the typical problems of Transformers, such as quadratic memory and time complexity. Furthermore, all prevalent architectures for PDE solving lack at least one of several desirable properties of an ideal surrogate model, such as (i) generalization to PDE parameters not seen during training, (ii) spatial and temporal zero-shot super-resolution, (iii) continuous temporal extrapolation, (iv) support for 1D, 2D, and 3D PDEs, and (v) efficient inference for longer temporal rollouts. To address these limitations, we propose Vectorized Conditional Neural Fields (VCNeFs), which represent the solution of time-dependent PDEs as neural fields. Contrary to prior methods, however, VCNeFs compute, for a set of multiple spatio-temporal query points, their solutions in parallel and model their dependencies through attention mechanisms. Moreover, VCNeF can condition the neural field on both the initial conditions and the parameters of the PDEs. An extensive set of experiments demonstrates that VCNeFs are competitive with and often outperform existing ML-based surrogate models.
format Preprint
id arxiv_https___arxiv_org_abs_2406_03919
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Vectorized Conditional Neural Fields: A Framework for Solving Time-dependent Parametric Partial Differential Equations
Hagnberger, Jan
Kalimuthu, Marimuthu
Musekamp, Daniel
Niepert, Mathias
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Neural and Evolutionary Computing
Computational Physics
Transformer models are increasingly used for solving Partial Differential Equations (PDEs). Several adaptations have been proposed, all of which suffer from the typical problems of Transformers, such as quadratic memory and time complexity. Furthermore, all prevalent architectures for PDE solving lack at least one of several desirable properties of an ideal surrogate model, such as (i) generalization to PDE parameters not seen during training, (ii) spatial and temporal zero-shot super-resolution, (iii) continuous temporal extrapolation, (iv) support for 1D, 2D, and 3D PDEs, and (v) efficient inference for longer temporal rollouts. To address these limitations, we propose Vectorized Conditional Neural Fields (VCNeFs), which represent the solution of time-dependent PDEs as neural fields. Contrary to prior methods, however, VCNeFs compute, for a set of multiple spatio-temporal query points, their solutions in parallel and model their dependencies through attention mechanisms. Moreover, VCNeF can condition the neural field on both the initial conditions and the parameters of the PDEs. An extensive set of experiments demonstrates that VCNeFs are competitive with and often outperform existing ML-based surrogate models.
title Vectorized Conditional Neural Fields: A Framework for Solving Time-dependent Parametric Partial Differential Equations
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Neural and Evolutionary Computing
Computational Physics
url https://arxiv.org/abs/2406.03919