November 11, 2022 Master Thesis Open Access

Wemmer, Florian

JSON Export

{
  "conceptrecid": "22141", 
  "created": "2022-12-02T16:21:49.589034+00:00", 
  "files": [
    {
      "bucket": "f5cb1b30-b42d-416c-b0ba-6dc08e97ed65", 
      "checksum": "md5:10467487be40bb285ef4c08e55f243d8", 
      "key": "Thesis.pdf", 
      "links": {
        "self": "https://publish.etp.kit.edu/api/files/f5cb1b30-b42d-416c-b0ba-6dc08e97ed65/Thesis.pdf"
      }, 
      "size": 24167209, 
      "type": "pdf"
    }
  ], 
  "id": 22142, 
  "links": {
    "bucket": "https://publish.etp.kit.edu/api/files/f5cb1b30-b42d-416c-b0ba-6dc08e97ed65", 
    "html": "https://publish.etp.kit.edu/record/22142", 
    "latest": "https://publish.etp.kit.edu/api/records/22142", 
    "latest_html": "https://publish.etp.kit.edu/record/22142"
  }, 
  "metadata": {
    "access_right": "open", 
    "access_right_category": "success", 
    "communities": [
      {
        "id": "belle2"
      }, 
      {
        "id": "etp"
      }
    ], 
    "contributors": [], 
    "creators": [
      {
        "affiliation": "KIT/ETP", 
        "name": "Wemmer, Florian"
      }
    ], 
    "description": "<p>This thesis presents the implementation and performance of the GravNet algorithm for the photon<br>\nenergy reconstruction in the Belle II electromagnetic calorimeter. GravNet is a machine learning<br>\nalgorithm based on the concept of graph neural networks. The Belle II Analysis Software Frame-<br>\nwork is the currently used reconstruction framework that serves as the baseline for comparison in<br>\nseveral studies. GravNet solves many of the conceptual restrictions that limit the performance<br>\nof the traditional reconstruction approach, especially in the presence of high levels of beam<br>\nbackground. The studies in this thesis are considered a first validation and are exclusively based<br>\non Monte Carlo generated and simulated data. The GravNet implementation outperforms the<br>\nbaseline energy resolutions over a large range of photon energies from 0.01GeV to 3.0 GeV by<br>\nup to 20 %. In addition, the studies demonstrate substantial improvements of up to 15 % in the<br>\nreconstruction of neutral pions from the invariant mass of two-photon systems. GravNet proves<br>\nto be a viable and versatile reconstruction algorithm with a promising outlook for a broad range<br>\nof present and future applications.</p>", 
    "keywords": [
      "Belle II", 
      "Calorimeter", 
      "Graph Neural Networks", 
      "Deep Learning", 
      "Clustering", 
      "Photon Reconstruction"
    ], 
    "language": "eng", 
    "publication_date": "2022-11-11", 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "22142"
          }, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "22141"
          }
        }
      ]
    }, 
    "resource_type": {
      "subtype": "master-thesis", 
      "title": "Master Thesis", 
      "type": "thesis"
    }, 
    "thesis": {
      "supervisors": [
        {
          "affiliation": "KIT/ETP", 
          "name": "Ferber, Torben"
        }, 
        {
          "affiliation": "KIT/ETP", 
          "name": "Klute, Markus"
        }
      ], 
      "university": "Karlsruhe Institute of Technology (KIT)"
    }, 
    "title": "Photon Reconstruction in the Belle II Calorimeter Using Graph Neural Networks"
  }, 
  "owners": [
    97
  ], 
  "revision": 3, 
  "stats": {}, 
  "updated": "2022-12-02T16:21:53.392818+00:00"
}