{
  "$schema": "./item.schema.json",
  "id": "pinns-neurips",
  "type": "publication",
  "cvs": [
    "np"
  ],
  "date": {
    "start": "2025-12"
  },
  "title": "Physics-Informed Neural Networks for Modeling Galactic Gravitational Potentials",
  "status": "published",
  "entryType": "inproceedings",
  "authors": [
    {
      "family": "Myers",
      "given": "Charlotte",
      "orcid": "0009-0005-9329-8196"
    },
    {
      "family": "Starkman",
      "given": "Nathaniel",
      "me": true,
      "orcid": "0000-0003-3954-3291"
    },
    {
      "family": "Necib",
      "given": "Lina",
      "orcid": "0000-0003-2806-1414"
    }
  ],
  "venue": {
    "booktitle": "Advances in Neural Information Processing Systems (NeurIPS 2025)"
  },
  "links": [
    {
      "rel": "paper",
      "url": "https://openreview.net/forum?id=JJQAd9eKwq",
      "label": "OpenReview"
    },
    {
      "rel": "code",
      "url": "https://github.com/charlottemyers/galactoPINNs"
    }
  ],
  "tags": [
    "machine-learning",
    "dynamics"
  ],
  "citekey": "Myers+:2025:galactic-pinns-neurips",
  "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\\% 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."
}