INEUS: Iterative Neural Solver for High-Dimensional PIDEs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dupret, Jean-Loup, Gallon, Davide, Cheridito, Patrick
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917468467888128
author Dupret, Jean-Loup
Gallon, Davide
Cheridito, Patrick
author_facet Dupret, Jean-Loup
Gallon, Davide
Cheridito, Patrick
contents In this paper, we introduce INEUS, a meshfree iterative neural solver for partial integro-differential equations (PIDEs). The method replaces the explicit evaluation of nonlocal jump integrals with single-jump sampling and reformulates PIDE solving as a sequence of recursive regression problems. Like Physics-Informed Neural Networks (PINNs), INEUS learns global solutions over the entire space-time domain, yet it offers a more efficient treatment of nonlocal terms and avoids the computationally expensive differentiation of full PIDE residuals. These features make INEUS particularly well suited for high-dimensional PDEs and PIDEs. Supported by a contraction-based convergence proof for linear PIDEs, our numerical experiments show that INEUS delivers accurate and scalable solutions for various high-dimensional linear and nonlinear examples.
format Preprint
id arxiv_https___arxiv_org_abs_2605_06281
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle INEUS: Iterative Neural Solver for High-Dimensional PIDEs
Dupret, Jean-Loup
Gallon, Davide
Cheridito, Patrick
Machine Learning
Numerical Analysis
Computational Finance
35R09, 65M99
In this paper, we introduce INEUS, a meshfree iterative neural solver for partial integro-differential equations (PIDEs). The method replaces the explicit evaluation of nonlocal jump integrals with single-jump sampling and reformulates PIDE solving as a sequence of recursive regression problems. Like Physics-Informed Neural Networks (PINNs), INEUS learns global solutions over the entire space-time domain, yet it offers a more efficient treatment of nonlocal terms and avoids the computationally expensive differentiation of full PIDE residuals. These features make INEUS particularly well suited for high-dimensional PDEs and PIDEs. Supported by a contraction-based convergence proof for linear PIDEs, our numerical experiments show that INEUS delivers accurate and scalable solutions for various high-dimensional linear and nonlinear examples.
title INEUS: Iterative Neural Solver for High-Dimensional PIDEs
topic Machine Learning
Numerical Analysis
Computational Finance
35R09, 65M99
url https://arxiv.org/abs/2605.06281