Search for effective field theory parameters for H→ZZ*→4ℓ using normalizing flow models

Abstract

The ATLAS and CMS collaborations announced the discovery of a new scalar particle of 125 GeV mass in 2012 whose measured properties like production cross-sections, couplings with other Standard Model (SM) particles, charge-parity are consistent with the predictions of SM within current measurement uncertainties. In the SM, interactions of the gauge bosons with Higgs boson lead to their masses via the mechanism of spontaneous symmetry breaking. However, the SM does not explain the existence of dark matter and dominance of matter over anti-matter. Also, mass of the Higgs boson gets large corrections from quantum fluctuations in the SM theory. In absence of any direct evidence of beyond SM (BSM) physics, an alternative approach is to investigate BSM interactions of Higgs boson modelled by an Effective Field Theory (EFT). An EFT incorporates higher dimensional interaction terms which could potentially modify the already observed SM interactions. This thesis aims to explore machine learning techniques like normalizing flows based on a real-valued non-volume preserving approach to search for effects of EFT interaction dubbed the cWW interaction on modifying the SM HZZ interactions in four lepton final state i.e. H→ZZ*→ℓ+ℓ-ℓ+ℓ- where ℓ could be electron or muon. The analysis is implemented as a parameter estimation using likelihoods for which a complete statistical analysis is performed for the ML model.