Skip to Main content Skip to Navigation

Beyond vanilla new physics at the LHC

Abstract : The ATLAS and CMS collaborations has put forward an extensive program of searches for new physics, aiming to cover new physics theories as much as possible. The results of this searches are typically interpreted in the context of popular models or simplified model topologies. However, there exist a plethora of BSM theories and scenarios that are not directly covered by the experimental analyses. Thus, the importance of reinterpreting LHC searches inthe context of any BSM theory. Is within this spirit that this thesis unfolds, by considering two aspects: phenomenology of beyond vanilla new physics theories and the further development of dedicated tools.On the phenomenological aspect, we explore the collider phenomenology of a very interesting beyond vanilla theory: The Minimal Dirac Gaugino Model (MDGSSM). First, we derive current limits on the gluinos and squarks of the model. Secondly, we start by delimiting the parameter space of the MGSSM electroweakino sector where the lightest neutralino is a viable DM candidate, to then constrain the emerging scenarios on the light of SUSY and LLP searches. On the tool development aspect, we describe recent developments of the SModelS reinterpretation tool, which includes combination of signal regions, the implementation of heavy stable charge particle and R-hadron signatures and the implementation of an interactive plots maker. Furthermore, we discuss the development of a tool to determine if signal regions from different analyses arestatistically independent from each other and hence, can be trivially combined.Finally, this thesis also regards machine learning (ML) applications to HEP phenomenology. Among other examples, we present a neural network to accurately predict, with an estimated uncertainty, the production cross sections of the Inert doublet Model. Furthermore, we discuss the importance of providing uncertainties on ML predictions and provide recommendations for sharing all the material involved in the production of ML applications.
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Wednesday, February 17, 2021 - 1:12:10 PM
Last modification on : Friday, March 25, 2022 - 9:42:44 AM
Long-term archiving on: : Tuesday, May 18, 2021 - 6:49:46 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03144142, version 1




Humberto Reyes Gonzalez. Beyond vanilla new physics at the LHC. High Energy Physics - Phenomenology [hep-ph]. Université Grenoble Alpes [2020-..], 2020. English. ⟨NNT : 2020GRALY037⟩. ⟨tel-03144142⟩



Record views


Files downloads