https://doi.org/10.1140/epjc/s10052-019-7562-1
Special Article - Tools for Experiment and Theory
DeepXS: fast approximation of MSSM electroweak cross sections at NLO
1
Institute for Mathematics, Astro- and Particle Physics IMAPP, Radboud Universiteit, Nijmegen, The Netherlands
2
GRAPPA, University of Amsterdam, Amsterdam, The Netherlands
3
Faculty of Physics, University of Warsaw, Warsaw, Poland
4
Nikhef, Amsterdam, The Netherlands
5
Mandelstam Institute for Theoretical Physics, University of the Witwatersrand, Johannesburg, South Africa
6
National Institute for Theoretical Physics, University of the Witwatersrand, Johannesburg, South Africa
7
Instituto de Fisica Corpuscular, IFIC-UV/CSIC University of Valencia, Valencia, Spain
8
Institute for Theoretical Particle Physics and Cosmology, RWTH Aachen University, Aachen, Germany
9
ESR Labs, Munich, Germany
* e-mail: Sydney.Otten@ru.nl
Received:
6
June
2019
Accepted:
16
December
2019
Published online:
7
January
2020
We present a deep learning solution to the prediction of particle production cross sections over a complicated, high-dimensional parameter space. We demonstrate the applicability by providing state-of-the-art predictions for the production of charginos and neutralinos at the Large Hadron Collider (LHC) at the next-to-leading order in the phenomenological MSSM-19 and explicitly demonstrate the performance for 
and 
as a proof of concept which will be extended to all SUSY electroweak pairs. We obtain errors that are lower than the uncertainty from scale and parton distribution functions with mean absolute percentage errors of well below 
allowing a safe inference at the next-to-leading order with inference times that improve the Monte Carlo integration procedures that have been available so far by a factor of 
from 
to 
per evaluation.
© The Author(s), 2020
