Speaker
Mr
Matthew Leigh
(University of Cape Town)
Description
The missing transverse momentum ($E_T^{miss}$) of a $pp$ collision is an important observable as it serves as a proxy for the transverse momentum carried away by undetectable particles. In addition to Standard Model neutrinos, there exist many theorised particles which would not leave any signals in a detector. Therefore, measuring $E_T^{miss}$ with high degrees of accuracy is critical for the understanding many physical processes which take place at the LHC, in the Standard Model and beyond. ATLAS currently utilises several working points to reconstruct the $E_T^{miss}$ of the hard-scatter interaction using only the observed objects. The optimal choice of working point depends strongly on the event topology. The aim of this project is to investigate the use of a deep neural network, trained using Monte-Carlo samples, to develop a more accurate $E_T^{miss}$ reconstruction, which is independent of topology and more resilient to pile-up effects. We demonstrate how the network improves $E_T^{miss}$ accuracy in both simulated and real data, and how this new method might assist in searches for physics beyond the Standard Model.
Apply to be<br> considered for a student <br> award (Yes / No)?
Yes
Level for award<br> (Hons, MSc, <br> PhD, N/A)?
MSc
Primary author
Mr
Matthew Leigh
(University of Cape Town)
Co-author
Dr
Sahal Yacoob
(University of Cape Town)
