PhD thesis defense
"On 22 December 2020 our colleague Giles Strong successfully defended his PhD thesis at IST. The research work was developed at LIP and supervised by Michele Gallinaro. The thesis was approved with the highest mark, Distinção e Louvor"
The thesis, entitled "Deep Learning methods applied to Higgs physics at the LHC", contains original work that includes development of innovative methods and advanced algorithms, and extended contributions to the analysis of LHC data with a focus on the study of Higgs boson properties. With the discovery of the Higgs boson in 2012, this area of research moved into the regime of precision measurements and experimental verification of theoretically predicted processes; any deviation from the Standard Model could be a sign of new physics and open new avenues for progressing our understanding of Nature.
The work of Giles focused on Machine Learning (ML) and Deep Learning (DL) methods and on their application to the search for double Higgs production in the final state with pairs of tau leptons and b-jets. The study was performed with the data collected in proton-proton collisions by the CMS experiment at CERN's LHC. Results considerably improve when these algorithms are applied to data analysis. The study of Higgs pair production will allow to probe the Higgs boson self-coupling parameters, and possibly detect coupling modifications. These studies directly help to shed light on the e electroweak symmetry breaking mechanism.
Giles developed and tested his innovative ML and DL algorithms in the context of the EU Horizon2020 funded AMVA4NewPhysics Marie Sklodowska-Curie innovative training network, in which he had a leading role.
