In this seminar, I shall present and discuss algorithmic approaches based on Machine Learning techniques to probe new physics at collider experiments, such as the LHC and future lepton machines. To find the best suite of neural networks that better adapt to distinct physics, I will discuss genetic algorithms, designed to optimize the various hyperparameters involved in the construction of the neural network architectures. We apply these tools to the study of a class of BSM models, focusing on theories containing vector-like fermions, both of the lepton and quark types, as well as multi-Higgs models with new neutral and charged scalars.
[The seminar will be recorded]
With support from FCT through project UIDB/00777/2020