July 22, 2024 Bachelor Thesis Open Access

Bogner, Lars

JSON-LD (schema.org) Export

{
  "@context": "https://schema.org/", 
  "@type": "ScholarlyArticle", 
  "contributor": [], 
  "creator": [
    {
      "@type": "Person", 
      "affiliation": "KIT/ETP", 
      "name": "Bogner, Lars"
    }
  ], 
  "datePublished": "2024-07-22", 
  "description": "<p>The next generation of e<sup>+</sup>e<sup>-</sup>-colliders, such as the proposed Future Circular Collider electron-positron (FCC-ee) at CERN operating as a Higgs and electroweak factory, necessitates new implementations while enabling innovative approaches in &tau; lepton reconstruction. In the FCC dataset, the &tau; lepton plays a crucial role in many physics topics. Focusing on the IDEA detector concept at the FCC-ee, machine learning (ML) approaches for jet flavor tagging are studied. Additionally, explicit reconstruction algorithms are employed for tagging the decay mode of &tau; leptons from its decay products on the jet level.</p>\n\n<p>In the study about using ML for identifying the flavor of Z boson decay jets, the existing, state-of-the-art Particle Transformer Network shows very good performance with very low false positive rates in quark jet suppression against Z &rarr; &tau;<sup>+</sup>&tau;<sup>-</sup> signal samples. A more simplistic feed-forward neural network is compared against the transformer model and shows worse performance with an increased false positive rate of two orders of magnitude.</p>\n\n<p><br>\nUsing an explicit reconstruction approach for the &tau; decay mode identification, the combination with the reconstruction of intermediate states, like the neutral &pi;<sup>0</sup> meson or the &rho;(770) meson is proposed. The performance of this method is evaluated in dependence on multiple cuts on the jet properties, aiming at optimizing the purity of the identified decay<br>\nmodes and investigating the efficiency-purity-relation. The proposed techniques promise good performance in &tau; reconstruction at FCC-ee but also further the need for a full simulation of the Innovative Detector for Electron-positron Accelerator (IDEA) detector to find final conclusions on the limitations of &tau; lepton reconstruction, as the current fast simulation shows limitations for the reconstruction of &tau; leptons. Additionally, alternative approaches to the tasks at hand are proposed.</p>", 
  "headline": "Tau Lepton Reconstruction at FCC-ee", 
  "image": "https://publish.etp.kit.edu/static/img/logos/zenodo-gradient-round.svg", 
  "inLanguage": {
    "@type": "Language", 
    "alternateName": "eng", 
    "name": "English"
  }, 
  "keywords": [
    "FCC-ee", 
    "machine learning", 
    "explicit reconstruction"
  ], 
  "name": "Tau Lepton Reconstruction at FCC-ee", 
  "url": "https://publish.etp.kit.edu/record/22243"
}