English

Physics-Informed Neural Networks for Modeling Galactic Gravitational Potentials

Astrophysics of Galaxies 2026-04-02 v2

Abstract

We introduce a physics-informed neural framework for modeling static and time-dependent galactic gravitational potentials. The method combines data-driven learning with embedded physical constraints to capture complex, small-scale features while preserving global physical consistency. We quantify predictive uncertainty through a Bayesian framework, and model time evolution using a neural ODE approach. Applied to mock systems of varying complexity, the model achieves reconstruction errors at the sub-percent level (0.14%0.14\% mean acceleration error) and improves dynamical consistency compared to analytic baselines. This method complements existing analytic methods, enabling physics-informed baseline potentials to be combined with neural residual fields to achieve both interpretable and accurate potential models.

Keywords

Cite

@article{arxiv.2602.01806,
  title  = {Physics-Informed Neural Networks for Modeling Galactic Gravitational Potentials},
  author = {Charlotte Myers and Nathaniel Starkman and Lina Necib},
  journal= {arXiv preprint arXiv:2602.01806},
  year   = {2026}
}

Comments

7 pages, 4 figures

R2 v1 2026-07-01T09:31:16.467Z