Accelerating Monte Carlo event generation -- rejection sampling using neural network event-weight estimates
arXiv:2109.11964 · doi:10.21468/SciPostPhys.12.5.164
Abstract
The generation of unit-weight events for complex scattering processes presents a severe challenge to modern Monte Carlo event generators. Even when using sophisticated phase-space sampling techniques adapted to the underlying transition matrix elements, the efficiency for generating unit-weight events from weighted samples can become a limiting factor in practical applications. Here we present a novel two-staged unweighting procedure that makes use of a neural-network surrogate for the full event weight. The algorithm can significantly accelerate the unweighting process, while it still guarantees unbiased sampling from the correct target distribution. We apply, validate and benchmark the new approach in high-multiplicity LHC production processes, including +4 jets and +3 jets, where we find speed-up factors up to ten.
arXiv comment37 pages, 9 figures, version accepted for publication in SciPost Physics
References in corpus (73)
- PYTHIA 6.4 Physics and Manual
- PYTHIA 6.2 Physics and Manual
- The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
- The anti-k_t jet clustering algorithm
- An Introduction to PYTHIA 8.2
- Parton distributions for the LHC
- A general framework for implementing NLO calculations in shower Monte Carlo programs: the POWHEG BOX
- Parton distributions for the LHC Run II
- Event generation with SHERPA 1.1
- HERWIG 6.5: an event generator for Hadron Emission Reactions With Interfering Gluons (including supersymmetric processes)
- ALPGEN, a generator for hard multiparton processes in hadronic collisions
- Matching NLO QCD computations and parton shower simulations
- The ATLAS Simulation Infrastructure
- A General Algorithm for Calculating Jet Cross Sections in NLO QCD
- NICE: Non-linear Independent Components Estimation
- Variational Inference with Normalizing Flows
- Scattering Amplitudes with Open Loops
- Normalizing Flows: An Introduction and Review of Current Methods
- Herwig 7.0 / Herwig++ 3.0 Release Note
- An update on vector boson pair production at hadron colliders
- Comix, a new matrix element generator
- Event Generation with Sherpa 2.2
- MadEvent: Automatic Event Generation with MadGraph
- QCD matrix elements + parton showers: The NLO case
- Three-jet cross sections to next-to-leading order
- WHIZARD: Simulating Multi-Particle Processes at LHC and ILC
- OpenLoops 2
- General-purpose event generators for LHC physics
- Automation of one-loop QCD corrections
- AMEGIC++ 1.0, A Matrix Element Generator In C++
- Robust Independent Validation of Experiment and Theory: Rivet version 3
- The automation of next-to-leading order electroweak calculations
- A posteriori inclusion of parton density functions in NLO QCD final-state calculations at hadron colliders: The APPLGRID Project
- RECOLA: REcursive Computation of One-Loop Amplitudes
- Recola2: REcursive Computation of One-Loop Amplitudes 2
- Weight optimization in multichannel Monte Carlo
- Event Generation with Normalizing Flows
- How to GAN LHC Events
- Numerical evaluation of virtual corrections to multi-jet production in massless QCD
- Vegas Revisited: Adaptive Monte Carlo Integration Beyond Factorization
- Automated Parton-Shower Variations in Pythia 8
- Reweighting QCD matrix-element and parton-shower calculations
- i-flow: High-dimensional Integration and Sampling with Normalizing Flows
- fastNLO: Fast pQCD Calculations for PDF Fits
- Exploring phase space with Neural Importance Sampling
- aMCfast: automation of fast NLO computations for PDF fits
- DijetGAN: A Generative-Adversarial Network Approach for the Simulation of QCD Dijet Events at the LHC
- On the maximal use of Monte Carlo samples: re-weighting events at NLO accuracy
- Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)
- Understanding Event-Generation Networks via Uncertainties
- Challenges in Monte Carlo event generator software for High-Luminosity LHC
- Neural Network-Based Approach to Phase Space Integration
- GANplifying Event Samples
- Simulation of vector boson plus many jet final states at the high luminosity LHC
- A factorisation-aware Matrix element emulator
- Reweighting Parton Showers
- Optimising simulations for diphoton production at hadron colliders using amplitude neural networks
- Foam: A General-Purpose Cellular Monte Carlo Event Generator
- Phase Space Sampling and Inference from Weighted Events with Autoregressive Flows
- PineAPPL: combining EW and QCD corrections for fast evaluation of LHC processes
- Foam: Multi-Dimensional General Purpose Monte Carlo Generator With Self-Adapting Simplical Grid
- Using neural networks for efficient evaluation of high multiplicity scattering amplitudes
- Improved Neural Network Monte Carlo Simulation
- DCTRGAN: Improving the Precision of Generative Models with Reweighting
- MCgrid: projecting cross section calculations on grids
- Efficient Monte Carlo Integration Using Boosted Decision Trees and Generative Deep Neural Networks
- Accelerating LHC phenomenology with analytic one-loop amplitudes: A C++ interface to MCFM
- Exhaustive Neural Importance Sampling applied to Monte Carlo event generation
- (MC)**3 -- a Multi-Channel Markov Chain Monte Carlo algorithm for phase-space sampling
- Matrix Element Regression with Deep Neural Networks -- breaking the CPU barrier
- Uncertainties associated with GAN-generated datasets in high energy physics
- : the cross-section evaluation code
- OASIS: Optimal Analysis-Specific Importance Sampling for event generation
Cited by in corpus (34)
- Machine Learning and LHC Event Generation
- Event Generators for High-Energy Physics Experiments
- Generative Networks for Precision Enthusiasts
- Targeting Multi-Loop Integrals with Neural Networks
- MadNIS -- Neural Multi-Channel Importance Sampling
- How to Understand Limitations of Generative Networks
- Two Invertible Networks for the Matrix Element Method
- The MadNIS Reloaded
- Modelling and computational improvements to the simulation of single vector-boson plus jet processes for the ATLAS experiment
- Efficient phase-space generation for hadron collider event simulation
- Accelerating LHC event generation with simplified pilot runs and fast PDFs
- One-loop matrix element emulation with factorisation awareness
- Unweighting multijet event generation using factorisation-aware neural networks
- CaloDREAM -- Detector Response Emulation via Attentive flow Matching
- Precision-Machine Learning for the Matrix Element Method
- ELSA -- Enhanced latent spaces for improved collider simulations
- Normalizing Flows for High-Dimensional Detector Simulations
- A Portable Parton-Level Event Generator for the High-Luminosity LHC
- Differentiable MadNIS-Lite
- Convolutional L2LFlows: Generating Accurate Showers in Highly Granular Calorimeters Using Convolutional Normalizing Flows
- Exploring phase space with Nested Sampling
- Accurate Surrogate Amplitudes with Calibrated Uncertainties
- Sampling NNLO QCD phase space with normalizing flows
- Machine Learning Post-Minkowskian Integrals
- Accelerating multijet-merged event generation with neural network matrix element surrogates
- Improving Colour Computations in MadGraph5_aMC@NLO and Exploring a 1/Nc Expansion
- Extrapolating Jet Radiation with Autoregressive Transformers
- Flow Annealed Importance Sampling Bootstrap meets Differentiable Particle Physics
- BitHEP -- The Limits of Low-Precision ML in HEP
- Amplitude Uncertainties Everywhere All at Once
- Amplitude Surrogates for Multi-Jet Processes
- How to Unfold Top Decays
- On the simulated kinematic distributions of heavy meson decays
- A Novel Implementation of the Matrix Element Method at Next-to-Leading Order for the Measurement of the Higgs Self-Coupling