Active Learning for Exclusion level set estimation with the ATLAS experiment

Jul 12, 2021, 3:30 PM
Track E (Zoom)

Track E


talk Computation, Machine Learning, and AI Computation, Machine Learning, and AI


Irina Espejo Morales (New York University (US))


Excursion is a tool to efficiently estimate level sets of
computationally expensive black box functions using Active Learning.
Excursion uses a Gaussian Process Regression as a surrogate model for
the black box function. It queries the target function (black box) iteratively in order to increase the available information regarding the desired level sets. We implement Excursion using GPyTorch which provides
state-of-the-art fast posterior fitting techniques and takes advantage
of GPUs to scale computations to higher dimensions.

In this talk, we demonstrate that Excursion significantly outperforms
traditional grid search approaches and we will detail the current work
in progress on improving Exotics searches as an intermediate step towards the ATLAS Run 2 pMSSM scan on $pp$ collisions at $\sqrt{s}=$ 13 TeV with the ATLAS detector.

Are you are a member of the APS Division of Particles and Fields? No

Primary author

Irina Espejo Morales (New York University (US))


Lukas Alexander Heinrich (CERN) Patrick Rieck (Max-Planck-Institut fur Physik (DE)) Prof. Gilles Louppe (University of Liège) Kyle Stuart Cranmer (New York University (US))

Presentation materials