Fourier Neural Operators for Non-Markovian Processes:Approximation Theorems and Experiments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lee, Wonjae, Kim, Taeyoung, Park, Hyungbin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916860885204992
author Lee, Wonjae
Kim, Taeyoung
Park, Hyungbin
author_facet Lee, Wonjae
Kim, Taeyoung
Park, Hyungbin
contents This paper introduces an operator-based neural network, the mirror-padded Fourier neural operator (MFNO), designed to learn the dynamics of stochastic systems. MFNO extends the standard Fourier neural operator (FNO) by incorporating mirror padding, enabling it to handle non-periodic inputs. We rigorously prove that MFNOs can approximate solutions of path-dependent stochastic differential equations and Lipschitz transformations of fractional Brownian motions to an arbitrary degree of accuracy. Our theoretical analysis builds on Wong--Zakai type theorems and various approximation techniques. Empirically, the MFNO exhibits strong resolution generalization--a property rarely seen in standard architectures such as LSTMs, TCNs, and DeepONet. Furthermore, our model achieves performance that is comparable or superior to these baselines while offering significantly faster sample path generation than classical numerical schemes.
format Preprint
id arxiv_https___arxiv_org_abs_2507_17887
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fourier Neural Operators for Non-Markovian Processes:Approximation Theorems and Experiments
Lee, Wonjae
Kim, Taeyoung
Park, Hyungbin
Machine Learning
Numerical Analysis
This paper introduces an operator-based neural network, the mirror-padded Fourier neural operator (MFNO), designed to learn the dynamics of stochastic systems. MFNO extends the standard Fourier neural operator (FNO) by incorporating mirror padding, enabling it to handle non-periodic inputs. We rigorously prove that MFNOs can approximate solutions of path-dependent stochastic differential equations and Lipschitz transformations of fractional Brownian motions to an arbitrary degree of accuracy. Our theoretical analysis builds on Wong--Zakai type theorems and various approximation techniques. Empirically, the MFNO exhibits strong resolution generalization--a property rarely seen in standard architectures such as LSTMs, TCNs, and DeepONet. Furthermore, our model achieves performance that is comparable or superior to these baselines while offering significantly faster sample path generation than classical numerical schemes.
title Fourier Neural Operators for Non-Markovian Processes:Approximation Theorems and Experiments
topic Machine Learning
Numerical Analysis
url https://arxiv.org/abs/2507.17887