Higgs boson pair production is a well-known probe of the structure of the electroweak symmetry breaking sector. We illustrate this using the gluon-fusion processes $pp\to H\to hh\to (b\overline{b})(b\overline{b})$ in the framework of two-Higgs-doublet models and show how a machine learning approach (three-stream convolutional neural network) can substantially improve the signal-background discrimination and thus improve the sensitivity coverage of the relevant parameter space. We further show that such $gg\to hh\to b\overline{b}b\overline{b}$ processes can probe the parameter space currently allowed by higgssignals and higgsbounds at the HL-LHC. Results are presented for 2HDM types I through IV.

• Received 16 August 2022
• Accepted 24 October 2022

DOI:https://doi.org/10.1103/PhysRevD.106.095015

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. Funded by SCOAP³.

Published by the American Physical Society