Two Invertible Networks for the Matrix Element Method

Abstract

The matrix element method is widely considered the ultimate LHC inference tool for small event numbers. We show how a combination of two conditional generative neural networks encodes the QCD radiation and detector effects without any simplifying assumptions, while keeping the computation of likelihoods for individual events numerically efficient. We illustrate our approach for the CP-violating phase of the top Yukawa coupling in associated Higgs and single-top production. Currently, the limiting factor for the precision of our approach is jet combinatorics.

arXiv comment25 pages, 13 figures

References in corpus (73)

• The anti-k_t jet clustering algorithm
• An Introduction to PYTHIA 8.2
• FastJet user manual
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• CaloGAN: Simulating 3D High Energy Particle Showers in Multi-Layer Electromagnetic Calorimeters with Generative Adversarial Networks
• Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis
• A Precision Measurement of the Mass of the Top Quark
• Guided Image Generation with Conditional Invertible Neural Networks
• A framework for Higgs characterisation
• A Roadmap for HEP Software and Computing R&D for the 2020s
• Event Generation with Normalizing Flows
• How to GAN LHC Events
• OmniFold: A Method to Simultaneously Unfold All Observables
• Constraining anomalous Higgs boson couplings to the heavy flavor fermions using matrix element techniques
• Getting High: High Fidelity Simulation of High Granularity Calorimeters with High Speed
• Precise simulation of electromagnetic calorimeter showers using a Wasserstein Generative Adversarial Network
• Machine Learning and LHC Event Generation
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• Automation of the matrix element reweighting method
• i-flow: High-dimensional Integration and Sampling with Normalizing Flows
• Boosting the Direct CP Measurement of the Higgs-Top Coupling
• Search for a standard model Higgs boson produced in association with a top-quark pair and decaying to bottom quarks using a matrix element method
• Exploring phase space with Neural Importance Sampling
• Measurement of the Top Quark Mass in the Dilepton Channel
• Invertible Networks or Partons to Detector and Back Again
• 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
• Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)
• Understanding Event-Generation Networks via Uncertainties
• The Matrix Element Method at Next-to-Leading Order
• Accelerating Monte Carlo event generation -- rejection sampling using neural network event-weight estimates
• Evidence for single top-quark production in the $s$-channel in proton-proton collisions at $\sqrt{s}=$8 TeV with the ATLAS detector using the Matrix Element Method
• Neural Network-Based Approach to Phase Space Integration
• GANplifying Event Samples
• Generative Networks for Precision Enthusiasts
• The Matrix Element Method and QCD Radiation
• The Matrix Element Method and its Application to Measurements of the Top Quark Mass
• Unravelling $t\bar{t}h$ via the matrix element method
• Unveiling CP property of top-Higgs coupling with graph neural networks at the LHC
• Deep-Learning Jets with Uncertainties and More
• Unbiased determination of the proton structure function F_2^p with faithful uncertainty estimation
• Targeting Multi-Loop Integrals with Neural Networks
• Phase Space Sampling and Inference from Weighted Events with Autoregressive Flows
• Extending the Matrix Element Method beyond the Born approximation: Calculating event weights at next-to-leading order accuracy
• Improved Neural Network Monte Carlo Simulation
• Per-Object Systematics using Deep-Learned Calibration
• Reweighting a parton shower using a neural network: the final-state case
• Polishing a shiny Higgs with matrix elements
• Optimized probes of $CP$-odd effects in the $t\bar{t}h$ process at hadron colliders
• Machine learning the Higgs boson-top quark CP phase
• Top Quark Mass Measurement from Dilepton Events at CDF II with the Matrix-Element Method
• Measuring QCD Splittings with Invertible Networks
• Indirect $\mathcal{C}\mathcal{P}$ probes of the Higgs-top-quark interaction: current LHC constraints and future opportunities
• Inference of cosmic-ray source properties by conditional invertible neural networks
• Measurement of spin correlations in t t-bar production using the matrix element method in the muon + jets final state in pp collisions at sqrt(s) = 8 TeV
• Towards a Computer Vision Particle Flow
• Single top-quark production at the Tevatron and the LHC
• Modern Machine Learning for LHC Physicists
• Probing the CP structure of the top quark Yukawa coupling: Loop sensitivity vs. on-shell sensitivity
• Direct Higgs-top CP-phase measurement with $t\bar{t}h$ at the 14 TeV LHC and 100 TeV FCC
• Measuring the Higgs-bottom coupling in weak boson fusion
• ν-Flows: Conditional Neutrino Regression
• Constraining CP-violation in the Higgs-top-quark interaction using machine-learning-based inference
• Finding the Higgs boson in decays to Z gamma using the matrix element method at Next-to-Leading Order
• The Matrix Element Method at next-to-leading order QCD for hadronic collisions: Single top-quark production at the LHC as an example application
• How to GAN Event Subtraction
• Matrix Element Method at NLO for (anti-)${\mathbf{k}}_{\mathbf{t}}$-jet algorithms
• Matrix Element Regression with Deep Neural Networks -- breaking the CPU barrier
• The matrix element method at next-to-leading order for arbitrary jet algorithms
• Reconstructing the invisible with matrix elements

Cited by in corpus (24)

• MadNIS -- Neural Multi-Channel Importance Sampling
• How to Understand Limitations of Generative Networks
• The MadNIS Reloaded
• Returning CP-Observables to The Frames They Belong
• CaloFlow for CaloChallenge Dataset 1
• CaloDREAM -- Detector Response Emulation via Attentive flow Matching
• Precision-Machine Learning for the Matrix Element Method
• A Lorentz-Equivariant Transformer for All of the LHC
• Normalizing Flows for High-Dimensional Detector Simulations
• Differentiable MadNIS-Lite
• Optimal, fast, and robust inference of reionization-era cosmology with the 21cmPIE-INN
• Anomaly detection with flow-based fast calorimeter simulators
• Accurate Surrogate Amplitudes with Calibrated Uncertainties
• Advancing Tools for Simulation-Based Inference
• PIPPIN: Generating variable length full events from partons
• Interpretable deep learning models for the inference and classification of LHC data
• Unifying Simulation and Inference with Normalizing Flows
• Generative Unfolding with Distribution Mapping
• Application-driven Validation of Posteriors in Inverse Problems
• Sensitivity to $\mathcal{C}\mathcal{P}$-violating effective couplings in the top-Higgs sector
• Extrapolating Jet Radiation with Autoregressive Transformers
• BitHEP -- The Limits of Low-Precision ML in HEP
• Reconstructing Sparticle masses at the LHC using Generative Machine Learning
• A Novel Implementation of the Matrix Element Method at Next-to-Leading Order for the Measurement of the Higgs Self-Coupling ${\lambda }_{3H}$