Novel b-jet Analyses for Beyond the Standard Model Higgs Bosons at the LHC

Ford, Billy, George (2021) Novel b-jet Analyses for Beyond the Standard Model Higgs Bosons at the LHC. University of Southampton, Doctoral Thesis, 134pp.

Record type: Thesis (Doctoral)

Abstract

Despite the many successes of the LHC since running began in 2008 - such as the discovery of the Higgs Boson in 2012 - the hunt for new physics beyond the Standard Model (SM) remains elusive. One particularly appealing extension to the SM comes in the form of Two-Higgs-Doublet-Models (2HDMs), which provide an enriched scalar sector of additional Higgs particles (the CP-even H and h, CP-odd A, and charged H^±). Where kinematically possible, interactions between these Higgses can lead to high b-jet multiplicity final states via H → hh → bb ^¯b ^¯b decays. In this thesis, various novel approaches to observing such new physics are considered. An alternative to traditional jet clustering algorithms, using a variable-R cone dependent on jet p_T , is shown to increase the potential signal significance when modelled against the leading backgrounds. Furthermore, the use of high level machine learning is investigated. By mapping p_T -weighted pixels in a detector into images, we build a convolutional neural network (CNN) to classify wide cone b-jets in signal events coming from 2HDM decays, against the leading backgrounds. Finally, we present a novel approach to jet reconstruction using spectral clustering machine learning techniques, and compare the performance with the currently well established methods in use at the LHC.