Neural Network-Based Approach to Phase Space Integration

Abstract

Monte Carlo methods are widely used in particle physics to integrate and sample probability distributions (differential cross sections or decay rates) on multi-dimensional phase spaces. We present a Neural Network (NN) algorithm optimized to perform this task. The algorithm has been applied to several examples of direct relevance for particle physics, including situations with non-trivial features such as sharp resonances and soft/collinear enhancements. Excellent performance has been demonstrated in all examples, with the properly trained NN achieving unweighting efficiencies of between 30% and 75%. In contrast to traditional Monte Carlo algorithms such as VEGAS, the NN-based approach does not require that the phase space coordinates be aligned with resonant or other features in the cross section.

13+2 pages, 9 figures. v2: Improved discussion, one new figure. No changes to physics results or conclusions. v3: Minor clarifications and improvements to figures, plus one new figure. No changes to results or conclusions. Now 18 pages, 11 figures

References in corpus (5)

• The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
• Event Generation with Normalizing Flows
• i-flow: High-dimensional Integration and Sampling with Normalizing Flows
• Exploring phase space with Neural Importance Sampling
• Reweighting a parton shower using a neural network: the final-state case

Cited by in corpus (38)

• Machine Learning and LHC Event Generation
• Invertible Networks or Partons to Detector and Back Again
• Understanding Event-Generation Networks via Uncertainties
• Challenges in Monte Carlo event generator software for High-Luminosity LHC
• Event Generators for High-Energy Physics Experiments
• Accelerating Monte Carlo event generation -- rejection sampling using neural network event-weight estimates
• Generative Networks for Precision Enthusiasts
• GANplifying Event Samples
• A factorisation-aware Matrix element emulator
• Optimising simulations for diphoton production at hadron colliders using amplitude neural networks
• Targeting Multi-Loop Integrals with Neural Networks
• Phase Space Sampling and Inference from Weighted Events with Autoregressive Flows
• MadNIS -- Neural Multi-Channel Importance Sampling
• Improved Neural Network Monte Carlo Simulation
• Two Invertible Networks for the Matrix Element Method
• Local Unitarity: a representation of differential cross-sections that is locally free of infrared singularities at any order
• The MadNIS Reloaded
• Efficient phase-space generation for hadron collider event simulation
• Unweighting multijet event generation using factorisation-aware neural networks
• One-loop matrix element emulation with factorisation awareness
• CaloDREAM -- Detector Response Emulation via Attentive flow Matching
• Multi-variable Integration with a Neural Network
• Matrix Element Regression with Deep Neural Networks -- breaking the CPU barrier
• Normalizing Flows for High-Dimensional Detector Simulations
• Differentiable MadNIS-Lite
• Convolutional L2LFlows: Generating Accurate Showers in Highly Granular Calorimeters Using Convolutional Normalizing Flows
• Exploring phase space with Nested Sampling
• Uncertainties associated with GAN-generated datasets in high energy physics
• OASIS: Optimal Analysis-Specific Importance Sampling for event generation
• Normalizing flows for microscopic many-body calculations: an application to the nuclear equation of state
• Machine Learning Post-Minkowskian Integrals
• How to GAN Event Unweighting
• Flow Annealed Importance Sampling Bootstrap meets Differentiable Particle Physics
• Variance Reduction via Simultaneous Importance Sampling and Control Variates Techniques Using Vegas
• BitHEP -- The Limits of Low-Precision ML in HEP
• Amplitude Surrogates for Multi-Jet Processes
• On the simulated kinematic distributions of heavy meson decays
• Generators and the (Accelerated) Future