NEOPIC: A Neural Operator Framework for Particle-based Kinetic Plasma Simulations

DOI10.48550/ARXIV.2610.40528MaRDI QIDQ7409084arXivFDO

Authors Sriramkrishnan Muralikrishnan, Jeffrey Kelling, Michael Bussmann, Robert Speck, Paul Gibbon, Chelsea Maria John

Publication date 29 October 2026

Published in arXiv

Tldr We introduce NEOPIC, a neural operator framework that learns field operators directly from particles, bypassing grids or trees. This plug-and-play field solver has representation equivalence with the continuous operator and error structure of standard Monte Carlo methods, and achieves stability beyond training time windows.

Full work available at URL https://doi.org/10.48550/ARXIV.2610.40528

Mathematics Subject Classification ID

Vlasov equations (35Q83) Monte Carlo methods (65C05) Probabilistic methods, particle methods, etc. for initial value and initial-boundary value problems involving PDEs (65M75) Artificial neural networks and deep learning (68T07) Statistical mechanics of plasmas (82D10)

arXiv classification

cs.LGmath.NAphysics.comp-phphysics.plasm-ph

This page was built for publication: NEOPIC: A Neural Operator Framework for Particle-based Kinetic Plasma Simulations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7409084)