CaloGAN: Simulating 3D High Energy Particle Showers in Multi-Layer Electromagnetic Calorimeters with Generative Adversarial Networks

Abstract

The precise modeling of subatomic particle interactions and propagation through matter is paramount for the advancement of nuclear and particle physics searches and precision measurements. The most computationally expensive step in the simulation pipeline of a typical experiment at the Large Hadron Collider (LHC) is the detailed modeling of the full complexity of physics processes that govern the motion and evolution of particle showers inside calorimeters. We introduce \textsc{CaloGAN}, a new fast simulation technique based on generative adversarial networks (GANs). We apply these neural networks to the modeling of electromagnetic showers in a longitudinally segmented calorimeter, and achieve speedup factors comparable to or better than existing full simulation techniques on CPU ($100\times$-$1000\times$) and even faster on GPU (up to $\sim 10^5\times$). There are still challenges for achieving precision across the entire phase space, but our solution can reproduce a variety of geometric shower shape properties of photons, positrons and charged pions. This represents a significant stepping stone toward a full neural network-based detector simulation that could save significant computing time and enable many analyses now and in the future.

14 pages, 4 tables, 13 figures; version accepted by Physical Review D (PRD)

References in corpus (16)

• Big-Bang Nucleosynthesis
• Parton distributions with LHC data
• The ATLAS Simulation Infrastructure
• Top++: a program for the calculation of the top-pair cross-section at hadron colliders
• Reconstruction of three-dimensional porous media using generative adversarial neural networks
• The CT10 NNLO Global Analysis of QCD
• Uncertainties on alpha_S in global PDF analyses and implications for predicted hadronic cross sections
• Electron performance measurements with the ATLAS detector using the 2010 LHC proton-proton collision data
• Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis
• Electron reconstruction and identification efficiency measurements with the ATLAS detector using the 2011 LHC proton-proton collision data
• Electron efficiency measurements with the ATLAS detector using 2012 LHC proton-proton collision data
• Generative Adversarial Networks recover features in astrophysical images of galaxies beyond the deconvolution limit
• Jet-Images: Computer Vision Inspired Techniques for Jet Tagging
• Measurement of the photon identification efficiencies with the ATLAS detector using LHC Run-1 data
• Measurement of the t t-bar production cross section using events in the e mu final state in pp collisions at sqrt(s) = 13 TeV
• Measurements of top-quark pair differential cross-sections in the $e\mu$ channel in $pp$ collisions at $\sqrt{s}=13$ TeV using the ATLAS detector

Cited by in corpus (163)

• Machine learning and the physical sciences
• Theory-guided Data Science: A New Paradigm for Scientific Discovery from Data
• Jet Substructure at the Large Hadron Collider: A Review of Recent Advances in Theory and Machine Learning
• Deep learning for determining a near-optimal topological design without any iteration
• Anomaly Detection with Density Estimation
• A Roadmap for HEP Software and Computing R&D for the 2020s
• 70 years of machine learning in geoscience in review
• Learning New Physics from a Machine
• How to GAN LHC Events
• Getting High: High Fidelity Simulation of High Granularity Calorimeters with High Speed
• StressGAN: A Generative Deep Learning Model for 2D Stress Distribution Prediction
• Learning representations of irregular particle-detector geometry with distance-weighted graph networks
• Pulling Out All the Tops with Computer Vision and Deep Learning
• Guiding New Physics Searches with Unsupervised Learning
• The Metric Space of Collider Events
• Pileup Mitigation with Machine Learning (PUMML)
• Precise simulation of electromagnetic calorimeter showers using a Wasserstein Generative Adversarial Network
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• CosmoGAN: creating high-fidelity weak lensing convergence maps using Generative Adversarial Networks
• AtlFast3: the next generation of fast simulation in ATLAS
• Machine Learning and LHC Event Generation
• Calorimetry with Deep Learning: Particle Simulation and Reconstruction for Collider Physics
• Learning to Classify from Impure Samples with High-Dimensional Data
• Regressive and generative neural networks for scalar field theory
• Score-based Generative Models for Calorimeter Shower Simulation
• Invertible Networks or Partons to Detector and Back Again
• Neural Networks for Full Phase-space Reweighting and Parameter Tuning
• CaloFlow: Fast and Accurate Generation of Calorimeter Showers with Normalizing Flows
• 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
• LaSDI: Parametric Latent Space Dynamics Identification
• Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)
• Understanding Event-Generation Networks via Uncertainties
• Applications and Techniques for Fast Machine Learning in Science
• Event Generators for High-Energy Physics Experiments
• Decoding Photons: Physics in the Latent Space of a BIB-AE Generative Network
• Modern Machine Learning and Particle Physics
• GANplifying Event Samples
• Generative Networks for Precision Enthusiasts
• A Neural Resampler for Monte Carlo Reweighting with Preserved Uncertainties
• A factorisation-aware Matrix element emulator
• Fast Point Cloud Generation with Diffusion Models in High Energy Physics
• Geometry Score: A Method For Comparing Generative Adversarial Networks
• FPGA-accelerated machine learning inference as a service for particle physics computing
• CaloClouds: Fast Geometry-Independent Highly-Granular Calorimeter Simulation
• Fast simulation of muons produced at the SHiP experiment using Generative Adversarial Networks
• Evaluating generative models in high energy physics
• An operational definition of quark and gluon jets
• A guide for deploying Deep Learning in LHC searches: How to achieve optimality and account for uncertainty
• Phase Space Sampling and Inference from Weighted Events with Autoregressive Flows
• Lund jet images from generative and cycle-consistent adversarial networks
• EPiC-GAN: Equivariant Point Cloud Generation for Particle Jets
• Complex-valued neural networks for machine learning on non-stationary physical data
• PC-JeDi: Diffusion for Particle Cloud Generation in High Energy Physics
• Calomplification -- The Power of Generative Calorimeter Models
• Image-based model parameter optimization using Model-Assisted Generative Adversarial Networks
• L2LFlows: Generating High-Fidelity 3D Calorimeter Images
• DCTRGAN: Improving the Precision of Generative Models with Reweighting
• Learning to Simulate High Energy Particle Collisions from Unlabeled Data
• How to Understand Limitations of Generative Networks
• Two Invertible Networks for the Matrix Element Method
• gLaSDI: Parametric Physics-informed Greedy Latent Space Dynamics Identification
• CaloMan: Fast generation of calorimeter showers with density estimation on learned manifolds
• Generative Adversarial Networks (GAN) for compact beam source modelling in Monte Carlo simulations
• Binary JUNIPR: an interpretable probabilistic model for discrimination
• Electromagnetic Showers Beyond Shower Shapes
• CaloScore v2: Single-shot Calorimeter Shower Simulation with Diffusion Models
• Deep generative models for fast photon shower simulation in ATLAS
• LHC analysis-specific datasets with Generative Adversarial Networks
• Towards a Deep Learning Model for Hadronization
• Inductive Simulation of Calorimeter Showers with Normalizing Flows
• Designing Accurate Emulators for Scientific Processes using Calibration-Driven Deep Models
• CaloFlow II: Even Faster and Still Accurate Generation of Calorimeter Showers with Normalizing Flows
• A single $T$-gate makes distribution learning hard
• The MadNIS Reloaded
• Machine Learning in High Energy Physics Community White Paper
• PC-Droid: Faster diffusion and improved quality for particle cloud generation
• Nanosecond machine learning event classification with boosted decision trees in FPGA for high energy physics
• Deep Generative Models for Detector Signature Simulation: A Taxonomic Review
• Lorentz group equivariant autoencoders
• Context-Enriched Identification of Particles with a Convolutional Network for Neutrino Events
• CaloFlow for CaloChallenge Dataset 1
• Dual-Parameterized Quantum Circuit GAN Model in High Energy Physics
• Ultra-High-Resolution Detector Simulation with Intra-Event Aware GAN and Self-Supervised Relational Reasoning
• How to GAN Event Subtraction
• SARM: Sparse Autoregressive Model for Scalable Generation of Sparse Images in Particle Physics
• GAN-based Matrix Factorization for Recommender Systems
• CaloDREAM -- Detector Response Emulation via Attentive flow Matching
• Particle-based Fast Jet Simulation at the LHC with Variational Autoencoders
• Precision-Machine Learning for the Matrix Element Method
• A deep learning method for the trajectory reconstruction of cosmic rays with the DAMPE mission
• Unsupervised Quantum Circuit Learning in High Energy Physics
• Classification and Recovery of Radio Signals from Cosmic Ray Induced Air Showers with Deep Learning
• Graph Generative Adversarial Networks for Sparse Data Generation in High Energy Physics
• A Lorentz-Equivariant Transformer for All of the LHC
• A high-granularity calorimeter insert based on SiPM-on-tile technology at the future Electron-Ion Collider
• Compressing PDF sets using generative adversarial networks
• Conditional Born machine for Monte Carlo event generation
• Normalizing Flows for High-Dimensional Detector Simulations
• Parametrizing the Detector Response with Neural Networks
• A Novel Scenario in the Semi-constrained NMSSM
• Calorimeter shower superresolution
• Parameter Estimation using Neural Networks in the Presence of Detector Effects
• Learning Multivariate New Physics
• Differentiable MadNIS-Lite
• Convolutional L2LFlows: Generating Accurate Showers in Highly Granular Calorimeters Using Convolutional Normalizing Flows
• Photon Reconstruction in the Belle II Calorimeter Using Graph Neural Networks
• CaloShowerGAN, a Generative Adversarial Networks model for fast calorimeter shower simulation
• On the impact of selected modern deep-learning techniques to the performance and celerity of classification models in an experimental high-energy physics use case
• CALPAGAN: Calorimetry for Particles using GANs
• Graph Generative Models for Fast Detector Simulations in High Energy Physics
• Hyperparameter Optimization of Generative Adversarial Network Models for High-Energy Physics Simulations
• Generalizing to new geometries with Geometry-Aware Autoregressive Models (GAAMs) for fast calorimeter simulation
• Uncertainties associated with GAN-generated datasets in high energy physics
• Energy Levels of One Dimensional Anharmonic Oscillator via Neural Networks
• CaloGraph: Graph-based diffusion model for fast shower generation in calorimeters with irregular geometry
• Physics-informed active learning with simultaneous weak-form latent space dynamics identification
• CaloChallenge 2022: A Community Challenge for Fast Calorimeter Simulation
• Anomaly detection with flow-based fast calorimeter simulators
• Accurate Surrogate Amplitudes with Calibrated Uncertainties
• Effectiveness of denoising diffusion probabilistic models for fast and high-fidelity whole-event simulation in high-energy heavy-ion experiments
• Physics Validation of Novel Convolutional 2D Architectures for Speeding Up High Energy Physics Simulations
• Quantum Generative Adversarial Networks in a Continuous-Variable Architecture to Simulate High Energy Physics Detectors
• OASIS: Optimal Analysis-Specific Importance Sampling for event generation
• Deep Learning Jet Substructure from Two-Particle Correlation
• PIPPIN: Generating variable length full events from partons
• A Data-driven Event Generator for Hadron Colliders using Wasserstein Generative Adversarial Network
• GAN with an Auxiliary Regressor for the Fast Simulation of the Electromagnetic Calorimeter Response
• Unsupervised and lightly supervised learning in particle physics
• Variational Autoencoders for Jet Simulation
• Set-Conditional Set Generation for Particle Physics
• Strategic Plan for a Scientific Software Innovation Institute (S2I2) for High Energy Physics
• How to GAN Event Unweighting
• Parnassus: An Automated Approach to Accurate, Precise, and Fast Detector Simulation and Reconstruction
• Unifying Simulation and Inference with Normalizing Flows
• SR-GAN for SR-gamma: super resolution of photon calorimeter images at collider experiments
• Leveraging Staggered Tessellation for Enhanced Spatial Resolution in High-Granularity Calorimeters
• LHC Hadronic Jet Generation Using Convolutional Variational Autoencoders with Normalizing Flows
• Top-philic Machine Learning
• Synthesis of pulses from particle detectors with a Generative Adversarial Network (GAN)
• The use of Generative Adversarial Networks to characterise new physics in multi-lepton final states at the LHC
• Differentiable Surrogate for Detector Simulation and Design with Diffusion Models
• Extrapolating Jet Radiation with Autoregressive Transformers
• Latent Space Refinement for Deep Generative Models
• The LHCb ultra-fast simulation option, Lamarr: design and validation
• BitHEP -- The Limits of Low-Precision ML in HEP
• Deep learning approaches for neural decoding: from CNNs to LSTMs and spikes to fMRI
• Exploring Generative Physics Models with Scientific Priors in Inertial Confinement Fusion
• Amplitude Uncertainties Everywhere All at Once
• Reconstructing Sparticle masses at the LHC using Generative Machine Learning
• Statistically Optimal Generative Modeling with Maximum Deviation from the Empirical Distribution
• CaloHadronic: a diffusion model for the generation of hadronic showers
• Choose Your Diffusion: Efficient and flexible ways to accelerate the diffusion model in fast high energy physics simulation
• Data driven background estimation in HEP using Generative Adversarial Networks
• Improved Surrogates in Inertial Confinement Fusion with Manifold and Cycle Consistencies
• Ultra Fast Calorimeter Simulation with Generative Machine Learning on FPGAs
• Implicit Quantile Networks For Emulation in Jet Physics
• CaloClouds3: Ultra-Fast Geometry-Independent Highly-Granular Calorimeter Simulation
• ScatterPrism: convergence for generative simulation and inverse problems in particle and nuclear physics
• Geometric Priors for Scientific Generative Models in Inertial Confinement Fusion
• ParaFlow: fast calorimeter simulations parameterized in upstream material configurations
• Ultra-Fast Generation of Air Shower Images for Imaging Air Cherenkov Telescopes using Generative Adversarial Networks