{
  "annotation_id": "fcba8b83-2cf2-4432-9959-ec81e802b7a9",
  "date_created": "2026-03-02T18:00:53.237000Z",
  "date_modified": "2026-03-02T18:00:53.237000Z",
  "file_hash": "8a8c607c9f049041aa69f406acb4600c5f7f2d9f5e444ea48d5805fc735a2da6",
  "private": false,
  "record": {
    "abstract": "A neural network method is developed to discriminate direct photons from the\nneutral pion background in the PHOS spectrometer of the ALICE experiment at the\nLHC collider. The neural net has been trained to distinguish different classes\nof events by analyzing the energy-profile tensor of a cluster in its eigen\nvector coordinate system. Monte-Carlo simulations show that this method\ndiminishes by an order of magnitude the probability of $\\pi^0$-meson\nmisidentification as a photon with respect to the direct photon identification\nefficiency in the energy range up to 120 GeV.",
    "arxiv_id": "physics/0410277",
    "authors": [
      "M. Yu. Bogolyubsky",
      "Yu. V. Kharlov",
      "S. A. Sadovsky"
    ],
    "categories": [
      "physics.ins-det",
      "physics.data-an"
    ],
    "title": "Direct Photon Identification with Artificial Neural Network in the   Photon Spectrometer PHOS",
    "url": "https://arxiv.org/abs/physics/0410277"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "996ed6dd-b5d5-4ff8-a554-762e9dc7c124",
    "id": "arXiv Dataset IDs",
    "type": "Model",
    "variant": "snapshot-2026-03-01",
    "version": "0.1.0"
  },
  "user_id": 1000002
}