A study of HH → WWγγ searches at future hadron colliders with different machine-learning approaches

• Jialiang Sun (Hamburg University (DE))

Measuring Higgs boson pair production is important because it allows for direct access to the Higgs trilinear coupling. In LHC run 3 and the HL-LHC, the most sensitive channels are bb\gamma\gamma, bb\tau\tau, and bbbb. However, for higher center-of-mass energies and integrated luminosities at future hadron colliders, it may be beneficial to consider not only these prominent channels but also rarer but cleaner decay channels that comprise letpons, photons, and missing transverse energy. In this talk, Higgs boson pair production in the HH →WW( e^+-\mu^-+2\nu) decay channel is investigated at future hadron colliders. Monte Carlo simulations were used at a center-of-mass energy of \sqrt{s} = 14 TeV, with detector effects modeled using Delphes with the CMS Phase-2 detector cards. To improve the event classification, machine learning techniques such as Boosted Decision Trees, Multilayer Perceptrons, and the Kolmogorov-Arnold Network are applied. Additionally, methods based on asymptotic formulae and Monte Carlo simulations for calculating the expected significance and upper limits are explored.