Heidelberg 2022 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 70: Experimental Methods (general) 3

T 70.3: Vortrag

Mittwoch, 23. März 2022, 16:45–17:00, T-H29

b-Tagging studies for the ATLAS experiment — •Eleonora Loiacono — Deutsches Elektronen-Synchrotron DESY, Hamburg and Zeuthen, Germany

At the LHC, among the different Higgs boson production mechanisms, the associated production with a vector boson VH → bb is considered as the golden channel for the measurement of the Higgs coupling to b-quarks, due to its high sensitivity. In the VH(bb) analysis, the available statistics of the simulated Monte Carlo (MC) samples is largely reduced by the b-tagging algorithms, especially for background processes, resulting in large statistical fluctuations in the MC templates. The first study that will be presented, is related to the reduction of this statistical uncertainty. In order to avoid being dominated by this un- certainty a technique based on Truth Tagging is performed, exploiting a Graph Neural Network to make an optimal use of multidimensional information associated to jets. The second study relies on improving of the connection between Flavour Tagging scale factors measured on data and those corresponding to MC to MC variations. b- tagging is very important for VH(bb) and the calibration procedures provide the use of the Scale Factors (SF). The so-called smoothing procedure is currently applied to these SF, which allows to have SF that are not dependent on the binning of the histograms used to do the calibration. In particular, there will be a focus on reviewing the currently smoothing procedure and also on including more statistical tests on procedure currently adopted for data SF.