{
  "annotation_id": "2c9c4295-47c6-4298-9c97-5863764955d7",
  "date_created": "2026-03-02T18:00:14.824000Z",
  "date_modified": "2026-03-02T18:00:14.824000Z",
  "file_hash": "3b6a1f413ac27052838af8fe01441490c28430510c60c16efcb17db6e1397657",
  "private": false,
  "record": {
    "abstract": "An accurate impact parameter determination in a heavy ion collision is\ncrucial for almost all further analysis. The capabilities of an artificial\nneural network are investigated to that respect. A novel input generation for\nthe network is proposed, namely the transverse and longitudinal momentum\ndistribution of all outgoing (or actually detectable) particles. The neural\nnetwork approach yields an improvement in performance of a factor of two as\ncompared to classical techniques. To achieve this improvement simple network\narchitectures and a 5 by 5 input grid in (p_t,p_z) space are sufficient.",
    "arxiv_id": "nucl-th/9601024",
    "authors": [
      "S. A. Bass",
      "A. Bischoff",
      "J. A. Maruhn",
      "H. Stoecker",
      "W. Greiner"
    ],
    "categories": [
      "nucl-th"
    ],
    "doi": "10.1103/PhysRevC.53.2358",
    "journal_ref": "Phys.Rev.C53:2358-2363,1996",
    "title": "Neural Networks for Impact Parameter Determination",
    "url": "https://arxiv.org/abs/nucl-th/9601024"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "b26d33a1-e8e0-42a7-8c26-f2623945a3c3",
    "id": "arXiv Dataset IDs",
    "type": "Model",
    "variant": "snapshot-2026-03-01",
    "version": "0.1.0"
  },
  "user_id": 1000002
}