A Neural Resampler for Monte Carlo Reweighting with Preserved Uncertainties

Abstract

Monte Carlo event generators are an essential tool for data analysis in collider physics. To include subleading quantum corrections, these generators often need to produce negative weight events, which leads to statistical dilution of the datasets and downstream computational costs for detector simulation. Building on the recent proposal of a positive resampler method to rebalance weights within histogram bins, we introduce neural resampling: an unbinned approach to Monte Carlo reweighting based on neural networks that scales well to high-dimensional and variable-dimensional phase space. We pay particular attention to preserving the statistical properties of the event sample, such that neural resampling not only maintains the mean value of any observable but also its Monte Carlo uncertainty. This uncertainty preservation scheme is general and can also be applied to binned (non-neural network) resampling. To illustrate our neural resampling approach, we present a case study from the Large Hadron Collider of top quark pair production at next-to-leading order matched to a parton shower.

arXiv comment12 pages, 3 figures, no negative weights; v2: to match journal version

References in corpus (74)

• Adam: A Method for Stochastic Optimization
• Auto-Encoding Variational Bayes
• 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
• An Introduction to PYTHIA 8.2
• FastJet user manual
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• A general framework for implementing NLO calculations in shower Monte Carlo programs: the POWHEG BOX
• Stochastic Backpropagation and Approximate Inference in Deep Generative Models
• Parton distributions with LHC data
• Matching NLO QCD computations and parton shower simulations
• Parton distributions from high-precision collider data
• Dispelling the N^3 myth for the Kt jet-finder
• Herwig 7.0 / Herwig++ 3.0 Release Note
• Event Generation with Sherpa 2.2
• Three-jet cross sections to next-to-leading order
• CutTools: a program implementing the OPP reduction method to compute one-loop amplitudes
• Automation of one-loop QCD corrections
• CaloGAN: Simulating 3D High Energy Particle Showers in Multi-Layer Electromagnetic Calorimeters with Generative Adversarial Networks
• OneLOop: for the evaluation of one-loop scalar functions
• A general approach to jet cross sections in QCD
• Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis
• Accelerating Science with Generative Adversarial Networks: An Application to 3D Particle Showers in Multi-Layer Calorimeters
• Energy Flow Networks: Deep Sets for Particle Jets
• A Roadmap for HEP Software and Computing R&D for the 2020s
• Parameterized Machine Learning for High-Energy Physics
• Automated one-loop calculations: a proof of concept
• Flow-based generative models for Markov chain Monte Carlo in lattice field theory
• Event Generation with Normalizing Flows
• Equivariant flow-based sampling for lattice gauge theory
• Ninja: Automated Integrand Reduction via Laurent Expansion for One-Loop Amplitudes
• How to GAN LHC Events
• OmniFold: A Method to Simultaneously Unfold All Observables
• Mining gold from implicit models to improve likelihood-free inference
• Constraining Effective Field Theories with Machine Learning
• Getting High: High Fidelity Simulation of High Granularity Calorimeters with High Speed
• Simulation Assisted Likelihood-free Anomaly Detection
• A Guide to Constraining Effective Field Theories with Machine Learning
• Automated Parton-Shower Variations in Pythia 8
• Approximating Likelihood Ratios with Calibrated Discriminative Classifiers
• Estimation of Thermodynamic Observables in Lattice Field Theories with Deep Generative Models
• Precise simulation of electromagnetic calorimeter showers using a Wasserstein Generative Adversarial Network
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• Reweighting QCD matrix-element and parton-shower calculations
• Calorimetry with Deep Learning: Particle Simulation and Reconstruction for Collider Physics
• i-flow: High-dimensional Integration and Sampling with Normalizing Flows
• Regressive and generative neural networks for scalar field theory
• Exploring phase space with Neural Importance Sampling
• Neural Networks for Full Phase-space Reweighting and Parameter Tuning
• Fast and accurate simulation of particle detectors using generative adversarial networks
• DijetGAN: A Generative-Adversarial Network Approach for the Simulation of QCD Dijet Events at the LHC
• How to GAN away Detector Effects
• Integrand reduction of one-loop scattering amplitudes through Laurent series expansion
• Unfolding with Generative Adversarial Networks
• A Positive-Weight Next-to-Leading-Order Monte Carlo for Z Pair Hadroproduction
• Controlling Physical Attributes in GAN-Accelerated Simulation of Electromagnetic Calorimeters
• Fast simulation of muons produced at the SHiP experiment using Generative Adversarial Networks
• A guide for deploying Deep Learning in LHC searches: How to achieve optimality and account for uncertainty
• Deep Learning as a Parton Shower
• Lund jet images from generative and cycle-consistent adversarial networks
• Generative Models for Fast Calorimeter Simulation.LHCb case
• On the reduction of negative weights in MC@NLO-type matching procedures
• Using neural networks for efficient evaluation of high multiplicity scattering amplitudes
• GANs for generating EFT models
• Binary JUNIPR: an interpretable probabilistic model for discrimination
• Efficient Monte Carlo Integration Using Boosted Decision Trees and Generative Deep Neural Networks
• LHC analysis-specific datasets with Generative Adversarial Networks
• Cherenkov Detectors Fast Simulation Using Neural Networks
• Particle Generative Adversarial Networks for full-event simulation at the LHC and their application to pileup description
• Machine Learning Templates for QCD Factorization in the Search for Physics Beyond the Standard Model
• Towards a new generation of parton densities with deep learning models
• Resonance Searches with Machine Learned Likelihood Ratios
• Efficiency Parameterization with Neural Networks

Cited by in corpus (34)

• Challenges in Monte Carlo event generator software for High-Luminosity LHC
• Event Generators for High-Energy Physics Experiments
• 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
• Presenting Unbinned Differential Cross Section Results
• Improved Neural Network Monte Carlo Simulation
• Vector Boson Scattering Processes: Status and Prospects
• SymmetryGAN: Symmetry Discovery with Deep Learning
• The role of colour flows in matrix element computations and Monte Carlo simulations
• Accelerating LHC event generation with simplified pilot runs and fast PDFs
• One-loop matrix element emulation with factorisation awareness
• Combine and Conquer: Event Reconstruction with Bayesian Ensemble Neural Networks
• Scaffolding Simulations with Deep Learning for High-dimensional Deconvolution
• A simultaneous unbinned differential cross section measurement of twenty-four $Z$+jets kinematic observables with the ATLAS detector
• Multi-variable Integration with a Neural Network
• Unbiased Elimination of Negative Weights in Monte Carlo Samples
• A CT18 global PDF fit at the leading order in QCD
• An implementation of neural simulation-based inference for parameter estimation in ATLAS
• Full Event Particle-Level Unfolding with Variable-Length Latent Variational Diffusion
• Uncertainties associated with GAN-generated datasets in high energy physics
• Logarithmically-accurate and positive-definite NLO shower matching
• Neural Conditional Reweighting
• Efficient negative-weight elimination in large high-multiplicity Monte Carlo event samples
• Reweighting simulated events using machine-learning techniques in the CMS experiment
• How to GAN Event Unweighting
• A Practical Guide to Unbinned Unfolding
• Speeding up MadGraph5_aMC@NLO
• Stay Positive: Neural Refinement of Sample Weights
• Monte Carlo Event Generators
• Reducing negative weights in Monte Carlo event generation with Sherpa
• Improving statistical precision in Monte Carlo samples with negative weights via reweighting and uncertainty quantification
• Generators and the (Accelerated) Future
• Reweighting and Analysing Event Generator Systematics by Neural Networks on High-Level Features