Understanding Event-Generation Networks via Uncertainties

Abstract

Following the growing success of generative neural networks in LHC simulations, the crucial question is how to control the networks and assign uncertainties to their event output. We show how Bayesian normalizing flow or invertible networks capture uncertainties from the training and turn them into an uncertainty on the event weight. Fundamentally, the interplay between density and uncertainty estimates indicates that these networks learn functions in analogy to parameter fits rather than binned event counts.

arXiv comment24 pages

References in corpus (50)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
• Generative Adversarial Networks: An Overview
• Variational Inference: A Review for Statisticians
• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• An Introduction to Variational Autoencoders
• Variational Inference with Normalizing Flows
• Normalizing Flows: An Introduction and Review of Current Methods
• Density estimation using Real NVP
• Parton distributions with QED corrections
• CaloGAN: Simulating 3D High Energy Particle Showers in Multi-Layer Electromagnetic Calorimeters with Generative Adversarial Networks
• Normalizing Flows for Probabilistic Modeling and Inference
• Anomaly Detection with Density Estimation
• 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
• Event Generation with Normalizing Flows
• How to GAN LHC Events
• Getting High: High Fidelity Simulation of High Granularity Calorimeters with High Speed
• Precise simulation of electromagnetic calorimeter showers using a Wasserstein Generative Adversarial Network
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• Calorimetry with Deep Learning: Particle Simulation and Reconstruction for Collider 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
• 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
• Unfolding with Generative Adversarial Networks
• Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)
• Decoding Photons: Physics in the Latent Space of a BIB-AE Generative Network
• Neural Network-Based Approach to Phase Space Integration
• GANplifying Event Samples
• Deep-Learning Jets with Uncertainties and More
• A guide for deploying Deep Learning in LHC searches: How to achieve optimality and account for uncertainty
• Deep Learning as a Parton Shower
• Phase Space Sampling and Inference from Weighted Events with Autoregressive Flows
• 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
• Per-Object Systematics using Deep-Learned Calibration
• Measuring QCD Splittings with Invertible Networks
• Towards a Computer Vision Particle Flow
• Flows for simultaneous manifold learning and density estimation
• LHC analysis-specific datasets with Generative Adversarial Networks
• Structured Variational Learning of Bayesian Neural Networks with Horseshoe Priors
• Combine and Conquer: Event Reconstruction with Bayesian Ensemble Neural Networks
• Machine Learning Templates for QCD Factorization in the Search for Physics Beyond the Standard Model
• How to GAN Event Subtraction
• Applying Bayesian Neural Networks to Event Reconstruction in Reactor Neutrino Experiments
• How to GAN Event Unweighting

Cited by in corpus (36)

• Machine Learning and LHC Event Generation
• CaloFlow: Fast and Accurate Generation of Calorimeter Showers with Normalizing Flows
• 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
• Optimising simulations for diphoton production at hadron colliders using amplitude neural networks
• Targeting Multi-Loop Integrals with Neural Networks
• Uncertainty Aware Learning for High Energy Physics
• MadNIS -- Neural Multi-Channel Importance Sampling
• Calomplification -- The Power of Generative Calorimeter Models
• Presenting Unbinned Differential Cross Section Results
• How to Understand Limitations of Generative Networks
• Two Invertible Networks for the Matrix Element Method
• What's Anomalous in LHC Jets?
• Returning CP-Observables to The Frames They Belong
• CaloFlow for CaloChallenge Dataset 1
• CaloDREAM -- Detector Response Emulation via Attentive flow Matching
• The Landscape of Unfolding with Machine Learning
• Ephemeral Learning -- Augmenting Triggers with Online-Trained Normalizing Flows
• Precision-Machine Learning for the Matrix Element Method
• Normalizing Flows for High-Dimensional Detector Simulations
• Active learning BSM parameter spaces
• Exploring phase space with Nested Sampling
• Accurate Surrogate Amplitudes with Calibrated Uncertainties
• Hypergraphs in LHC Phenomenology -- The Next Frontier of IRC-Safe Feature Extraction
• Anomaly detection with flow-based fast calorimeter simulators
• Advancing Tools for Simulation-Based Inference
• Calibrating Bayesian Generative Machine Learning for Bayesiamplification
• Generative Unfolding with Distribution Mapping
• Unifying Simulation and Inference with Normalizing Flows
• Top-philic Machine Learning
• Latent Space Refinement for Deep Generative Models
• Extrapolating Jet Radiation with Autoregressive Transformers
• Observable Optimization for Precision Theory: Machine Learning Energy Correlators
• Event Generation and Density Estimation with Surjective Normalizing Flows
• Encoding off-shell effects in top pair production in Direct Diffusion networks