Higgs Boson Cross Section Measurement in the W-Associated Production and Fully Leptonic Final Stat at $\sqrt{s}$ = 13 $\mathrm{TeV}$ with the ATLAS Experiment using Deep Neural Networks

In this thesis, a neural network-based multivariate analysis is presented for the cross section measurement of the associated Higgs production channel \begin{align*} pp\rightarrow W \rightarrow WH \rightarrow WWW^*\rightarrow 3\ell~3\nu \end{align*} with the Higgs boson decaying into two \W~bosons and a fully leptonic final state. The analysis focuses on the most sensitive signal region of a leptonic final state without lepton pairs of the same lepton flavour and different electric charge. The data set that is used comprises 139 $\mathrm{fb}^{-1}$ of proton-proton collisions collected at the ATLAS detector at a center of mass energy of $\sqrt{s}$ = 13 $\mathrm{TeV}$. In the analysis a multiclass deep neural network is employed to suppress the most dominant backgrounds from top and $WZ$~decays. The expected significance is measured to be $Z_0$ = 2.93.