{
  "$schema": "./item.schema.json",
  "id": "galactopinns",
  "type": "software",
  "tier": "other",
  "cvs": [],
  "date": {
    "start": "2025-11",
    "present": true
  },
  "title": "galactoPINNs",
  "summary": "Physics-informed neural networks for galactic potentials",
  "highlight": {
    "topic": "software",
    "image": "highlights/pinns-mnras.webp",
    "alt": "Orbits of a test particle in a Milky Way–LMC system, integrated in the true potential, a near-true analytic model, and three neural-network models. The most complete neural-network models track the true orbit almost exactly, while the near-true analytic model drifts away from it.",
    "description": "Physics-informed neural networks for galactic potentials, learning only the corrections to an analytic model from stellar kinematics."
  },
  "repo": "charlottemyers/galactoPINNs",
  "links": [
    {
      "rel": "code",
      "url": "https://github.com/charlottemyers/galactoPINNs"
    }
  ],
  "tags": [
    "jax",
    "machine-learning"
  ],
  "refs": [
    "pinns-mnras"
  ]
}