November 29, 2021 to December 3, 2021
Virtual and IBS Science Culture Center, Daejeon, South Korea
Asia/Seoul timezone

Optimising simulations for diphoton production at hadron colliders using amplitude neural networks

contribution ID 567
Dec 1, 2021, 5:00 PM
S305 (Virtual and IBS Science Culture Center)


Virtual and IBS Science Culture Center

55 EXPO-ro Yuseong-gu Daejeon, South Korea email: +82 42 878 8299
Oral Track 3: Computations in Theoretical Physics: Techniques and Methods Track 3: Computations in Theoretical Physics: Techniques and Methods


Ryan Moodie (IPPP, Durham University)


Phenomenological studies of high-multiplicity scattering processes at collider experiments present a substantial theoretical challenge and are increasingly important ingredients in experimental measurements. We investigate the use of neural networks to approximate matrix elements for these processes, studying the case of loop-induced diphoton production through gluon fusion. We train neural network models on one-loop amplitudes from the NJet library and interface them with the Sherpa Monte Carlo event generator to provide the matrix element within a realistic hadronic collider simulation. Computing some standard observables with the models and comparing to conventional techniques, we find excellent agreement in the distributions and a reduced simulation time by a factor of 30.


We extend previous work which pioneered the emulation of scattering amplitudes with neural networks, studying these techniques for the first time within a full hadronic collider simulation.


Speaker time zone Compatible with Europe

Primary authors

Joseph Bullock (IPPP, Durham University) Ryan Moodie (IPPP, Durham University) Simon Badger (Università degli Studi di Torino)

Presentation materials