Publications

Order by: publication date number of citations

• Benchmarking LHC searches for light 2HDM+a pseudoscalars

  Spyros Argyropoulos, Ulrich Haisch

  SciPost Phys. 13, 007 (2022) · published 22 July 2022 | Toggle abstract · pdf

  Using two suitable benchmark scenarios that satisfy the experimental con- straints on the total decay width of the 125 GeV Higgs boson, we determine the bounds on light CP-odd spin-0 states in the 2HDM+a model that arise from existing LHC searches. Our work should prove useful for 2HDM+a interpretations of future ATLAS, CMS and LHCb searches for pseudoscalars with masses below the electroweak scale.

  6 citations

• Understanding Event-Generation Networks via Uncertainties

  Marco Bellagente, Manuel Haußmann, Michel Luchmann, Tilman Plehn

  SciPost Phys. 13, 003 (2022) · published 21 July 2022 | Toggle abstract · pdf

  Following the growing success of generative neural networks in LHC simulations, the crucial question is how to control the networks and assign uncertainties to their event output. We show how Bayesian normalizing flow or invertible networks capture uncertainties from the training and turn them into an uncertainty on the event weight. Fundamentally, the interplay between density and uncertainty estimates indicates that these networks learn functions in analogy to parameter fits rather than binned event counts.

  19 citations

• Picking the low-hanging fruit: testing new physics at scale with active learning

  Juan Rocamonde, Louie Corpe, Gustavs Zilgalvis, Maria Avramidou, Jon Butterworth

  SciPost Phys. 13, 002 (2022) · published 15 July 2022 | Toggle abstract · pdf

  Since the discovery of the Higgs boson, testing the many possible extensions to the Standard Model has become a key challenge in particle physics. This paper discusses a new method for predicting the compatibility of new physics theories with existing experimental data from particle colliders. By using machine learning, the technique obtained comparable results to previous methods (>90% precision and recall) with only a fraction of their computing resources (<10%). This makes it possible to test models that were impossible to probe before, and allows for large-scale testing of new physics theories.

  4 citations