Study of the production of two same-charge polarised W bosons via vector boson scattering with neural networks

The Standard Model of particle physics (SM) predicts in what way elementary particles can interact. One such interaction is vector boson scattering (VBS) of 𝑊±𝑊±𝑗𝑗. In studying this VBS process we are able to probe electroweak symmetry breaking (EWSB) as a key feature of the SM directly. EWSB is how the W boson acquires mass and a longitudinal polarisation component. In the SM fine-tuned interferences prevent VBS processes from violating unitarity. This study uses Monte Carlo (MC) simulations of proton-proton collisions with VBS signature as input data for neural networks (NNs). The goal is to discriminate between VBS events with two longitudinally-polarised W bosons, 𝑊±𝐿 𝑊±𝐿 , from 𝑊±𝑊±𝑗𝑗 events where at least one of the W bosons is transversely polarised. While validating the MC data, tests on different observable distributions indicate inconsistencies towards the expected behaviour, which need to be further investigated. This work presents different NN-topologies which are able to distinguish well between the signal and background processes. In further research the idealistic dataset of only 𝑊±𝑊±𝑗𝑗 events needs to be extended by more background processes and detector simulation for the NNs to work with experimental data.