Speaker
Description
We present an application of the gradient boosting machine learning technique for tagging top quark and W jets in hadronic four-top final states induced by both Standard Model (SM) and Beyond the Standard Model (BSM) processes. Our approach utilizes classical subjettiness variables within the framework of the Delphes common parameterized detector simulation package. Jets produced in simulated proton-proton collisions at sqrt(s) = 14 TeV are identified as consistent with the hypothesis of originating from the decay of a top quark or a W boson and are used to reconstruct the mass of a hypothetical scalar resonance decaying into a pair of top quarks in events where four top quarks are produced. To evaluate the performance of our machine learning-based approach, we compare the results with those obtained using a simple cut-based tagging technique. Stacked histograms of a mixture of SM and BSM processes are analyzed, and the mass peak of the scalar resonance within the four-top final state is fitted to evaluate the performance of the ML and cut-based approaches. The application of these ML techniques enhances our ability to discern top quarks and W jets, contributing to a better understanding of the underlying physics processes. The results provide valuable insights for future experimental analyses, enabling the exploration of both SM and BSM physics scenarios.
