Search for Higgs boson pair production with the ATLAS detector at the LHC

Novak, Jakob (author), Kerševan, Borut Paul (mentor), Gorišek, Andrej (co-mentor)

The thesis presents the search for Higgs boson pair production where one Higgs boson decays via $H\to bb$ and the other Higgs boson decays via $H/S \to WW^* \to \ell\nu qq$ (where $\ell$ is either an electron or a muon). The latter can be Standard Model Higgs boson or beyond Standard Model Higgs boson with same couplings but different mass. Analysed data consists of pp collisions at a centre-of-mass energy of 13 TeV recorded with the ATLAS detector at the Large Hadron Collider. Full Run 2 dataset corresponding to integrated luminosity of 140.1 fb$^{-1}$ is used. Machine learning approach is adopted for selection of events matching the signal topology. No statistically significant evidence of a signal is observed and corresponding upper limits on the production cross-section for $HH$ and $SH$ production are derived.

• Language: English
• Keywords: ATLAS, machine learning, Higgs pair production, single-lepton, CERN, neural networks, Higgs self-interaction, two b-jets, LHC Run 2, classification problem, BSM scalar resonance, missing transverse momentum
• Work type: Doctoral dissertation
• Typology: 2.08 - Doctoral Dissertation
• Organization: FMF - Faculty of Mathematics and Physics
• Year: 2024
• PID: 20.500.12556/RUL-156036
• COBISS.SI-ID: 189458947
• Publication date in RUL: 01.05.2024

Secondary language

• Language: Slovenian
• Title: Iskanje procesov s parom Higgsovih bozonov z detektorjem ATLAS na trkalniku LHC

Abstract

V doktorskem delu sem predstavil iskanje procesov s parom Higgsovih bozonov, v katerih en Higgsov bozon razpade preko $H\to bb$ in drugi Higgsov bozon prek $H/S \to WW^* \to \ell\nu qq$ (kjer je $\ell$ bodisi elektron ali mion). Slednji je lahko Higgsov bozon Standardnega modela ali Higgsov bozon izven Standardnega modela, z enakimi sklopitvami, vendar drugačno maso. Ta analiza vključuje celotno statistiko drugega obdobja (Run 2) zajemanja podatkov z detektorjem ATLAS na trkalniku LHC. Integrirana luminoznost znaša 140,1 fb$^{-1}$ pri težiščni energiji 13 TeV. Za izbiro dogodkov, ki ustrezajo topologiji signala, je uporabljena metoda strojnega učenja. Opaženih ni nobenih statistično signifikantnih odstopanj od napovedi ozadja. Izračunane so izključitvene meje na sipalni presek za produkcijo $HH$ in $SH$ signala.

• Keywords: ATLAS, strojno učenje, pari Higgsovih bozonov, en rekonstruiran lepton, CERN, nevronske mreže, interakcija Higgsa s samim seboj, dva pljuska b, LHC Run 2, klasifikacijski problem, skalarna resonanca, manjkajoča transverzalna gibalna količina