Auricchio, Silvia (2022) Improving the search of new resonances Y decaying into a Higgs boson and a generic new particle X in hadronic final states with Machine Learning at the ATLAS detector. [Tesi di dottorato]
Preview |
Text
Auricchio_Silvia_35.pdf Download (40MB) | Preview |
| Item Type: | Tesi di dottorato |
|---|---|
| Resource language: | English |
| Title: | Improving the search of new resonances Y decaying into a Higgs boson and a generic new particle X in hadronic final states with Machine Learning at the ATLAS detector |
| Creators: | Creators Email Auricchio, Silvia silvia.auricchio@unina.it |
| Date: | 13 December 2022 |
| Number of Pages: | 247 |
| Institution: | Università degli Studi di Napoli Federico II |
| Department: | Fisica |
| Dottorato: | Fisica |
| Ciclo di dottorato: | 35 |
| Coordinatore del Corso di dottorato: | nome email Canale, Vincenzo canale@na.infn.it |
| Tutor: | nome email Rossi, Elvira UNSPECIFIED Conventi, Francesco UNSPECIFIED Merola, Leonardo UNSPECIFIED |
| Date: | 13 December 2022 |
| Number of Pages: | 247 |
| Keywords: | High Energy Physics; Machine Learning; Data analysis; Background estimation; Anomaly detection; LHC; ATLAS; CERN; Beyond Standard Model |
| Settori scientifico-disciplinari del MIUR: | Area 02 - Scienze fisiche > FIS/04 - Fisica nucleare e subnucleare |
| Date Deposited: | 20 Dec 2022 14:16 |
| Last Modified: | 09 Apr 2025 14:19 |
| URI: | http://www.fedoa.unina.it/id/eprint/14635 |
Collection description
High Energy Physics is now accessing a new era. The research for Physics Beyond Standard Model requires more and more data to prove the existence of eventual rare phenomena. This involves a big challenge on the processing of the considerable amount and complexity of data collected and Machine learning (ML) provides a valid solution. The work presented here is about the search for a heavy resonance Y decaying into a Standard Model Higgs boson H and a new particle X in a fully hadronic final state, with several ML applications which helped to improve several aspects of the analysis. The results presented are based on the 139 fb−1 dataset of proton-proton collisions at √s = 13 TeV, collected by the ATLAS detector from 2015 to 2018, during LHC Run-2. A novel anomaly detection signal region is implemented based on a jet-level score for signal model-independent tagging of the boosted X, representing the first application of fully unsupervised ML to an ATLAS analysis. Two additional signal regions are implemented to target a benchmark X decay to two quarks, covering topologies where the X is reconstructed as either a single large-radius jet or two small-radius jets. The Higgs is assumed to decay to b ̄b and its boosted topology is recognized among Quantum Chromodynamics and Top jets using a new ML-based tagger. The background estimation is totally data-driven and performed with the help of a Deep Neural Network. No significant excess of data is observed over the expected background, and the results are interpreted in upper limits at 95% confidence level on the production cross section σ(pp → Y → XH → q ̄qb ̄b) for signals with mY between 1.5 and 6 TeV and mX between 65 and 3000 GeV.
Downloads
Downloads per month over past year
Actions (login required)
 |
View Item |
