Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 47: Higgs Physics V

T 47.5: Vortrag

Mittwoch, 18. März 2026, 17:15–17:30, KH 00.014

Neural network classifier strategy for optimal Higgs boson self-coupling sensitivity in the CMS HH → bbττ analysis — Ana Andrade, •Benjamin Le, Bogdan Wiederspan, Marcel Rieger, Moritz Jonas Wolf, Nathan Prouvost, Peter Scleper, and Tobias Kramer — University of Hamburg

The Standard Model (SM) of particle physics remains one of the most accurate theories describing the universes matter and its fundamental interactions at the smallest scales. One prediction that is yet to be fully tested is the self-interaction of the Higgs boson, characterized by the trilinear coupling strength λ, which gives rise to the shape of the Higgs potential. Typical analysis strategies involve neural networks for signal-background classification, often trained with simulated signal events following the SM prediction for λ. However, in case the actual value of λ deviates from the SM expectation, kinematic properties are subject to change and therefore, rendering the choice of λ used during training suboptimal. This talk summarizes a study that addresses this challenge by exploring different neural network strategies, enhancing the sensitivity to a wide range of hypothetical self-coupling values.

Keywords: Machine learning; di-Higgs; bbtautau