Speaker
Description
Monte Carlo simulations of scattering processes with many particles require enormous computing power. Particularly in view of the HL-LHC, an improvement in efficiency is necessary in order to be able to carry out the desired investigations in an economically sensible way. We show that employing a sophisticated neural network emulation of QCD multijet matrix elements based on dipole factorisation can lead to a drastic acceleration of unweighted event generation in high-multiplicity LHC production processes. We incorporate these emulations as fast and accurate surrogates in a two-stage rejection sampling algorithm within the SHERPA Monte Carlo event generator. The approach reduces the computational cost of unweighted events by factors between 16 and 350 for the considered channels.
SciPost Phys. 15, 107 (2023), arXiv:2301.13562
| Alternate track | 14. Computing, AI and Data Handling |
|---|---|
| I read the instructions above | Yes |
