Doctoral Dissertation Open Access

Classical and Quantum Machine Learning for Anomaly Detection at the CMS Experiment and Beyond

Bal, Aritra

Citation Style Language JSON Export

{
  "abstract": "<p>This thesis presents two novel methodologies for anomaly detection in particle collision events, utilising both classical and quantum machine learning techniques. The first approach employs an unsupervised, data-driven variational autoencoder (VAE), trained on collision data recorded by the CMS detector at the Large Hadron Collider during Run 2.&nbsp;No significant excess above the Standard Model (SM) expectation is observed. This method is subsequently used to derive exclusion limits on a large number of exotic signal models derived from Beyond the Standard Model (BSM) physics.&nbsp;The second approach introduces a novel quantum machine learning framework, named <em>1P1Q</em>, for encoding jet kinematic features onto two-level quantum bits (qubits).&nbsp;For both the anomaly detection and supervised classification tasks, the <em>1P1Q</em> method is demonstrated to achieve signal identification and separation performance comparable to or surpassing current state-of-the-art classical machine learning algorithms.</p>", 
  "author": [
    {
      "family": "Bal, Aritra"
    }
  ], 
  "id": "22337", 
  "issued": {
    "date-parts": [
      [
        2025, 
        7, 
        11
      ]
    ]
  }, 
  "language": "eng", 
  "title": "Classical and Quantum Machine Learning for Anomaly Detection at the CMS Experiment and Beyond", 
  "type": "thesis"
}
Publication date:
July 11, 2025
Keyword(s):
machine learning quantum machine learning anomaly detection BSM physics EXO CMS model-agnostic unsupervised ML searches CMS ETP
Awarding University:
Karlsruhe Institute of Technology (KIT)
Report Number:
ETP-KA/2025-06
Communities:

Cite as

Export