Guardado en:
Detalles Bibliográficos
Autores principales: Kreider, Max, Thomas, Peter J., Li, Yao
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2508.20339
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909757173923840
author Kreider, Max
Thomas, Peter J.
Li, Yao
author_facet Kreider, Max
Thomas, Peter J.
Li, Yao
contents For a stochastic differential equation (SDE) that is an Itô diffusion or Langevin equation, the Fokker-Planck operator governs the evolution of the probability density, while its adjoint, the infinitesimal generator of the stochastic Koopman operator, governs the evolution of system observables, in the mean. The eigenfunctions of these operators provide a powerful framework to analyze SDEs, and have shown to be particularly useful for systems of stochastic oscillators. However, computing these eigenfunctions typically requires solving high-dimensional PDEs on unbounded domains, which is numerically challenging. Building on previous work, we propose a data-driven artificial neural network solver for Koopman and Fokker-Planck eigenfunctions. Our approach incorporates the differential operator into the loss function, improving accuracy and reducing dependence on large amounts of accurate training data. We demonstrate our approach on several numerical examples in two, three, and four dimensions.
format Preprint
id arxiv_https___arxiv_org_abs_2508_20339
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Artificial neural network solver for Fokker-Planck and Koopman eigenfunctions
Kreider, Max
Thomas, Peter J.
Li, Yao
Numerical Analysis
For a stochastic differential equation (SDE) that is an Itô diffusion or Langevin equation, the Fokker-Planck operator governs the evolution of the probability density, while its adjoint, the infinitesimal generator of the stochastic Koopman operator, governs the evolution of system observables, in the mean. The eigenfunctions of these operators provide a powerful framework to analyze SDEs, and have shown to be particularly useful for systems of stochastic oscillators. However, computing these eigenfunctions typically requires solving high-dimensional PDEs on unbounded domains, which is numerically challenging. Building on previous work, we propose a data-driven artificial neural network solver for Koopman and Fokker-Planck eigenfunctions. Our approach incorporates the differential operator into the loss function, improving accuracy and reducing dependence on large amounts of accurate training data. We demonstrate our approach on several numerical examples in two, three, and four dimensions.
title Artificial neural network solver for Fokker-Planck and Koopman eigenfunctions
topic Numerical Analysis
url https://arxiv.org/abs/2508.20339