Targeting Multi-Loop Integrals with Neural Networks

Abstract

Numerical evaluations of Feynman integrals often proceed via a deformation of the integration contour into the complex plane. While valid contours are easy to construct, the numerical precision for a multi-loop integral can depend critically on the chosen contour. We present methods to optimize this contour using a combination of optimized, global complex shifts and a normalizing flow. They can lead to a significant gain in precision.

arXiv comment20 pages, 9 figures, v3: added two references

References in corpus (48)

• TensorFlow: A system for large-scale machine learning
• Parton distributions for the LHC Run II
• NICE: Non-linear Independent Components Estimation
• Variational Inference with Normalizing Flows
• Normalizing Flows: An Introduction and Review of Current Methods
• An automatized algorithm to compute infrared divergent multi-loop integrals
• pySecDec: a toolbox for the numerical evaluation of multi-scale integrals
• Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis
• Guided Image Generation with Conditional Invertible Neural Networks
• Massive Feynman diagrams and inverse binomial sums
• SecDec-3.0: numerical evaluation of multi-scale integrals beyond one loop
• Event Generation with Normalizing Flows
• An algebraic/numerical formalism for one-loop multi-leg amplitudes
• Sector Decomposition
• How to GAN LHC Events
• Two-loop planar master integrals for Higgs$\to 3$ partons with full heavy-quark mass dependence
• Collider Physics at the Precision Frontier
• Numerical integration of one-loop Feynman diagrams for N-photon amplitudes
• A GPU compatible quasi-Monte Carlo integrator interfaced to pySecDec
• Techniques for QCD calculations by numerical integration
• Numerical evaluation of multi-loop integrals for arbitrary kinematics with SecDec 2.0
• QCD corrections to tri-boson production
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• i-flow: High-dimensional Integration and Sampling with Normalizing Flows
• Exploring phase space with Neural Importance Sampling
• Invertible Networks or Partons to Detector and Back Again
• DijetGAN: A Generative-Adversarial Network Approach for the Simulation of QCD Dijet Events at the LHC
• How to GAN away Detector Effects
• Master Integrals for the 2-loop QCD virtual corrections to the Forward-Backward Asymmetry
• Unfolding with Generative Adversarial Networks
• Evaluating multi-loop Feynman diagrams with infrared and threshold singularities numerically
• Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)
• Understanding Event-Generation Networks via Uncertainties
• Accelerating Monte Carlo event generation -- rejection sampling using neural network event-weight estimates
• Neural Network-Based Approach to Phase Space Integration
• Generative Networks for Precision Enthusiasts
• 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
• The Differential equation method: calculation of vertex-type diagrams with one non-zero mass
• Improved Neural Network Monte Carlo Simulation
• Reweighting a parton shower using a neural network: the final-state case
• Using neural networks for efficient evaluation of high multiplicity scattering amplitudes
• DCTRGAN: Improving the Precision of Generative Models with Reweighting
• Towards a Computer Vision Particle Flow
• The complete set of two-loop master integrals for Higgs + jet production in QCD
• How to GAN Event Subtraction
• Evaluation of multi-loop multi-scale integrals and phenomenological two-loop applications

Cited by in corpus (29)

• Machine Learning and LHC Event Generation
• Event Generators for High-Energy Physics Experiments
• Evaluation of Feynman integrals with arbitrary complex masses via series expansions
• MadNIS -- Neural Multi-Channel Importance Sampling
• Les Houches 2021: Physics at TeV Colliders: Report on the Standard Model Precision Wishlist
• Two Invertible Networks for the Matrix Element Method
• Quantum algorithm for Feynman loop integrals
• The MadNIS Reloaded
• CaloFlow for CaloChallenge Dataset 1
• Ephemeral Learning -- Augmenting Triggers with Online-Trained Normalizing Flows
• Comparing Machine Learning and Interpolation Methods for Loop-Level Calculations
• Precision-Machine Learning for the Matrix Element Method
• ELSA -- Enhanced latent spaces for improved collider simulations
• Calorimeter shower superresolution
• Differentiable MadNIS-Lite
• Accurate Surrogate Amplitudes with Calibrated Uncertainties
• Anomaly detection with flow-based fast calorimeter simulators
• PIPPIN: Generating variable length full events from partons
• Learning Feynman integrals from differential equations with neural networks
• Sampling NNLO QCD phase space with normalizing flows
• Machine Learning Post-Minkowskian Integrals
• Learning the Simplicity of Scattering Amplitudes
• TopicFlow: Disentangling quark and gluon jets with normalizing flows
• Unifying Simulation and Inference with Normalizing Flows
• Amplitude Uncertainties Everywhere All at Once
• Extrapolating Jet Radiation with Autoregressive Transformers
• BitHEP -- The Limits of Low-Precision ML in HEP
• Event Generation and Density Estimation with Surjective Normalizing Flows
• Amplitude Surrogates for Multi-Jet Processes