Searches for CP-Violation in $ H \rightarrow \tau \tau $ Decay with the ATLAS Detector

This thesis presents a measurement of the $CP$-properties of the Yukawa coupling between the Higgs boson and $\tau{\rm-lepton}$. The measurement uses the proton–proton collision data collected from 2015 to 2018 with the ATLAS detector at the Large Hadron Collider. Totally $139$~fb$^{-1}$ proton–proton collision data is collected at a center-of-mass energy of $\sqrt{s} = 13$~TeV. This study investigates the $CP$-properties with $CP$-sensitive observables defined by the visible decay products of $\tau{\rm-leptons}$. $CP$-violating interactions between the Higgs boson and $\tau{\rm-lepton}$ are described by the $CP$-mixing angle $\phi_{\tau}$. The expected value of $\phi_{\tau}$ according to the Standard Model is $0^\circ\pm28^\circ$ at $68\%$ confidence level, and $0^{+75\circ}_{-70}$ at $95.5\%$ confidence level obtained from a simulated dataset. The observed value of $\phi_{\tau}$ is $9^\circ\pm16^\circ$ at $68\%$ confidence level, and $9^\circ\pm34^\circ$ at $95.5\%$ confidence level. The pure $CP$-odd hypothesis is disfavoured at 3.4 standard deviations. The observation is consistent with the Standard Model expectations. This thesis also presents studies of pile-up jet tagging. The pile-up jet tagging algorithm is based on the K-Nearest Neighbor method. It is trained with the data collected from 2015 to 2018 with the ATLAS detector. A new pile-up jet tagging algorithm based on neural network is developed. It provides improvement to the background rejection at all working points. In this thesis, a new $CP$-sensitive observable is developed using neural network approach. The trained observable excludes the pure $CP$-odd hypothesis at higher confidence level in a simplified sample compared with the previously designed observable. These studies will contribute to future measurements of the $CP$-properties.