Accelerating multijet-merged event generation with neural network matrix element surrogates
arXiv:2506.06203 · doi:10.21468/SciPostPhys.20.3.071
Abstract
The efficient simulation of multijet final states presents a serious computational task for analyses of LHC data and will be even more so at the HL-LHC. We here discuss means to accelerate the generation of unweighted events based on a two-stage rejection-sampling algorithm that employs neural-network surrogates for unweighting the hard-process matrix elements. To this end, we generalise the previously proposed algorithm based on factorisation-aware neural networks to the case of multijet merging at tree-level accuracy. We thereby account for several non-trivial aspects of realistic event-simulation setups, including biased phase-space sampling, partial unweighting, and the mapping of partonic subprocesses. We apply our methods to the production of Z+jets final states at the HL-LHC using the Sherpa event generator, including matrix elements with up to six final-state partons. When using neural-network surrogates for the dominant Z+5 jets and Z+6 jets partonic processes, we find a reduction in the total event-generation time by more than a factor of 10 compared to baseline Sherpa.
37 pages, 10 figures
References in corpus (35)
- A parton shower algorithm based on Catani-Seymour dipole factorisation
- Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
- Event Generation with Sherpa 2.2
- Robust Independent Validation of Experiment and Theory: Rivet version 3
- Event Generation with Normalizing Flows
- Helac-Phegas: a generator for all parton level processes
- i-flow: High-dimensional Integration and Sampling with Normalizing Flows
- Exploring phase space with Neural Importance Sampling
- Event Generators for High-Energy Physics Experiments
- Accelerating Monte Carlo event generation -- rejection sampling using neural network event-weight estimates
- Simulation of vector boson plus many jet final states at the high luminosity LHC
- A factorisation-aware Matrix element emulator
- Optimising simulations for diphoton production at hadron colliders using amplitude neural networks
- Phase Space Sampling and Inference from Weighted Events with Autoregressive Flows
- MadNIS -- Neural Multi-Channel Importance Sampling
- Using neural networks for efficient evaluation of high multiplicity scattering amplitudes
- The MadNIS Reloaded
- Modelling and computational improvements to the simulation of single vector-boson plus jet processes for the ATLAS experiment
- Unweighting multijet event generation using factorisation-aware neural networks
- (MC)**3 -- a Multi-Channel Markov Chain Monte Carlo algorithm for phase-space sampling
- One-loop matrix element emulation with factorisation awareness
- A Lorentz-Equivariant Transformer for All of the LHC
- A Portable Parton-Level Event Generator for the High-Luminosity LHC
- Loop Amplitudes from Precision Networks
- Exploring phase space with Nested Sampling
- Accurate Surrogate Amplitudes with Calibrated Uncertainties
- Sampling NNLO QCD phase space with normalizing flows
- Accelerating HEP simulations with Neural Importance Sampling
- Event generation with Sherpa 3
- Madgraph5_aMC@NLO on GPUs and vector CPUs Experience with the first alpha release
- Efficient precision simulation of processes with many-jet final states at the LHC
- Many-gluon tree amplitudes on modern GPUs: A case study for novel event generators
- Flow Annealed Importance Sampling Bootstrap meets Differentiable Particle Physics
- Efficient Monte Carlo Event Generation for Neutrino-Nucleus Exclusive Cross Sections
- HEPScore: A new CPU benchmark for the WLCG