Event
High Energy Theory Seminar: Search for exotic Higgs boson decays with CMS and fast machine learning solutions for the LHC
Ho Fung Tsoi (University of Wisconsin-Madison)
There is still room for BSM physics in the scalar sector, which could manifest as Higgs boson decays to a pair of light pseudoscalars. I will present a search for such decay in final states with two b quarks and two muons or two tau leptons, using the LHC Run 2 data of proton-proton collisions collected by the CMS detector. On enhancing the overall experimental sensitivity at a lower level, I will introduce a novel machine learning-based trigger algorithm that uses anomaly detection technique to search for new physics in a model-agnostic way as close to the raw collision data as possible, i.e. the CMS Level-1 trigger, which filters collision events of 40 MHz at real time at a latency of O(100) ns on FPGAs. Symbolic regression for further accelerating machine learning inference at nanoseconds on FPGAs is also discussed.