Work

Embedding physical symmetries into machine-learned reduced plasma physics models via data augmentation

Year: 2026

Type: article

Abstract: Machine learning is offering powerful tools for the development and discovery of reduced models of nonlinear, multiscale plasma dynamics from the data of first-principles kinetic simulations. However, ensuring the physical consistency of such models requires embedding fundamental symmetries of plasma dynamics. In this work, we explore a symmetry-embedding strategy based on data augmentation, where symmetry-preserving transformations (e.g., Lorentz and Galilean boosts) are applied to simulation data. Using both sparse regression and neural networks, we show that models trained on symmetry-augmented data more accurately infer the plasma fluid equations and pressure tensor closures from fully kinetic particle-in-cell simulations of magnetic reconnection. We show that this approach suppresses spurious inertial-frame-dependent correlations between dynamical variables, improves data efficiency, and significantly outperforms models trained without symmetry-augmented data, as well as commonly used theoretical pressure closure models. Our results establish symmetry-based data augmentation as a broadly applicable method for incorporating physical structure into machine-learned reduced plasma models.

Source: Physical Review Research

DOI: 10.1103/9fq1-vpqz

Authors: M. C. McGrae-Menge, J. R. Pierce, F. Fiuza, E. P. Alves

Institutions: University of California, Los Angeles, Instituto Superior Técnico

Language: English

Topic: Model Reduction and Neural Networks

Subfield: Statistical and Nonlinear Physics

Field: Physics and Astronomy

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $2,625

Funders: National Science Foundation, National Energy Research Scientific Computing Center, European Research Council