i-flow: High-dimensional Integration and Sampling with Normalizing Flows

Abstract

In many fields of science, high-dimensional integration is required. Numerical methods have been developed to evaluate these complex integrals. We introduce the code i-flow, a python package that performs high-dimensional numerical integration utilizing normalizing flows. Normalizing flows are machine-learned, bijective mappings between two distributions. i-flow can also be used to sample random points according to complicated distributions in high dimensions. We compare i-flow to other algorithms for high-dimensional numerical integration and show that i-flow outperforms them for high dimensional correlated integrals. The i-flow code is publicly available on gitlab at https://gitlab.com/i-flow/i-flow.

21 pages, 5 figures, 4 tables; v2: improved presentation and discussion, matches published version. Mach. Learn.: Sci. Technol (2020)

References in corpus (27)

• Adam: A Method for Stochastic Optimization
• Automatized One-Loop Calculations in 4 and D dimensions
• NICE: Non-linear Independent Components Estimation
• Variational Inference with Normalizing Flows
• Density estimation using Real NVP
• Training Deep Nets with Sublinear Memory Cost
• General-purpose event generators for LHC physics
• Event Generation with Normalizing Flows
• How to GAN LHC Events
• Exploring phase space with Neural Importance Sampling
• On the Chi square and higher-order Chi distances for approximating f-divergences
• A-NICE-MC: Adversarial Training for MCMC
• Simulation of vector boson plus many jet final states at the high luminosity LHC
• NeuTra-lizing Bad Geometry in Hamiltonian Monte Carlo Using Neural Transport
• A numerical evaluation of the scalar hexagon integral in the physical region
• Differential Higgs production at N3LO beyond threshold
• Fast simulation of muons produced at the SHiP experiment using Generative Adversarial Networks
• Foam: A General-Purpose Cellular Monte Carlo Event Generator
• Learning Likelihoods with Conditional Normalizing Flows
• Lund jet images from generative and cycle-consistent adversarial networks
• Discrete Flows: Invertible Generative Models of Discrete Data
• Event Generation and Statistical Sampling for Physics with Deep Generative Models and a Density Information Buffer
• Efficient Monte Carlo Integration Using Boosted Decision Trees and Generative Deep Neural Networks
• LHC analysis-specific datasets with Generative Adversarial Networks
• Generalizing Hamiltonian Monte Carlo with Neural Networks
• Uncertainties associated with GAN-generated datasets in high energy physics
• Transflow Learning: Repurposing Flow Models Without Retraining

Cited by in corpus (68)

• Machine Learning and LHC Event Generation
• Invertible Networks or Partons to Detector and Back Again
• CaloFlow: Fast and Accurate Generation of Calorimeter Showers with Normalizing Flows
• Understanding Event-Generation Networks via Uncertainties
• Challenges in Monte Carlo event generator software for High-Luminosity LHC
• Accelerating Monte Carlo event generation -- rejection sampling using neural network event-weight estimates
• Event Generators for High-Energy Physics Experiments
• Neural Network-Based Approach to Phase Space Integration
• Generative Networks for Precision Enthusiasts
• GANplifying Event Samples
• A Neural Resampler for Monte Carlo Reweighting with Preserved Uncertainties
• 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
• Quasi Anomalous Knowledge: Searching for new physics with embedded knowledge
• MadNIS -- Neural Multi-Channel Importance Sampling
• Improved Neural Network Monte Carlo Simulation
• How to Understand Limitations of Generative Networks
• 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
• Measuring QCD Splittings with Invertible Networks
• Vector Boson Scattering Processes: Status and Prospects
• Towards a Deep Learning Model for Hadronization
• The MadNIS Reloaded
• Efficient phase-space generation for hadron collider event simulation
• CaloFlow for CaloChallenge Dataset 1
• CaloDREAM -- Detector Response Emulation via Attentive flow Matching
• Precision-Machine Learning for the Matrix Element Method
• Fourier-Flow model generating Feynman paths
• Comparing Machine Learning and Interpolation Methods for Loop-Level Calculations
• Multi-variable Integration with a Neural Network
• Matrix Element Regression with Deep Neural Networks -- breaking the CPU barrier
• Unbiased Elimination of Negative Weights in Monte Carlo Samples
• Normalizing Flows for High-Dimensional Detector Simulations
• Calorimeter shower superresolution
• 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
• Fully probabilistic quasar continua predictions near Lyman-α with conditional neural spline flows
• Exploring phase space with Nested Sampling
• Comparison of Affine and Rational Quadratic Spline Coupling and Autoregressive Flows through Robust Statistical Tests
• Uncertainties associated with GAN-generated datasets in high energy physics
• Accurate Surrogate Amplitudes with Calibrated Uncertainties
• Anomaly detection with flow-based fast calorimeter simulators
• Normalizing flows for microscopic many-body calculations: an application to the nuclear equation of state
• Sampling NNLO QCD phase space with normalizing flows
• OASIS: Optimal Analysis-Specific Importance Sampling for event generation
• Machine Learning Post-Minkowskian Integrals
• Efficient negative-weight elimination in large high-multiplicity Monte Carlo event samples
• Path Integral Sampler: a stochastic control approach for sampling
• TopicFlow: Disentangling quark and gluon jets with normalizing flows
• Accelerating multijet-merged event generation with neural network matrix element surrogates
• How to GAN Event Unweighting
• Annealed Flow Transport Monte Carlo
• Unifying Simulation and Inference with Normalizing Flows
• Variance Reduction via Simultaneous Importance Sampling and Control Variates Techniques Using Vegas
• Flow Annealed Importance Sampling Bootstrap meets Differentiable Particle Physics
• Extrapolating Jet Radiation with Autoregressive Transformers
• BitHEP -- The Limits of Low-Precision ML in HEP
• Amplitude Uncertainties Everywhere All at Once
• Latent Space Refinement for Deep Generative Models
• Amplitude Surrogates for Multi-Jet Processes
• Event Generation and Density Estimation with Surjective Normalizing Flows
• Observable Optimization for Precision Theory: Machine Learning Energy Correlators
• Implicit Quantile Networks For Emulation in Jet Physics
• On the simulated kinematic distributions of heavy meson decays
• Data-Driven Predictions for Dark Photon and Millicharged Particle Production
• Generators and the (Accelerated) Future