Searches for Higgs pair production with the ATLAS Experiment at the LHC

Description

Following the momentous discovery of the Higgs boson at the Large Hadron Collider by the ATLAS and CMS collaborations in 2012, substantial progress has been made in measuring its properties, including mass, spin and couplings to other particles. However, the ultimate test of electroweak symmetry breaking (ESWB) can only be obtained by reconstructing the characteristic Mexican-hat-shaped potential. Measurement of the Higgs self-coupling via production of Higgs boson pairs provides a unique direct probe of this potential, giving a key handle on many of the big open questions in the Standard Model. Given the di-Higgs cross-section is more than 1000 times smaller than that of single Higgs production, the full LHC dataset coupled with state-of-the art machine learning (ML) techniques will be crucial.  

The ATLAS group at Liverpool has played a leading role in searching for di-Higgs production since the beginning of LHC run-2, pioneering the most sensitive channel, where one of the Higgs bosons decays to tau-leptons and the other decays to b-jets (HH --> bbττ). The successful applicant will play a major role in this analysis with the full LHC dataset, along with the combination with other decay channels in ATLAS and with the corresponding CMS searches, to obtain first evidence for this process or stringent limits on any potential new physics enhancements, leading to several legacy LHC publications.  

Liverpool are playing major roles in the identification of the constituent tau-leptons and b-jets using graph neutral networks (GNNs) and their calibration. The PhD student would join one of these areas to optimise the performance. Within the analysis itself, Liverpool are studying the use of GNNs or transformer architectures to optimally distinguish signal from background and ML regression methods to improve the H --> tautau mass reconstruction, correcting for the energy lost as neutrinos. We also lead the study of the main top-pair background. The student would take a major role in one or more of these areas. 

The PhD position is part of the STFC-funded LIV.INNO Centre for Doctoral training, which will provide targeted data-science training throughout the duration that will be directly applied to the above analysis. The duration of the PhD will be 48 months and will be based primarily at the University of Liverpool with the opportunity to spend a year at CERN and frequent short trips to CERN. As part of the LIV.INNO scheme there is a mandatory six-month industry placement as part of the PhD. 

There will also be the possibility to attend at least one relevant training school and at least one international conference. 

Please ensure you include the project title and reference number PPPR070 when applying.