E Pluribus Unum Ex Machina: Learning from Many Collider Events at Once

Abstract

There have been a number of recent proposals to enhance the performance of machine learning strategies for collider physics by combining many distinct events into a single ensemble feature. To evaluate the efficacy of these proposals, we study the connection between single-event classifiers and multi-event classifiers under the assumption that collider events are independent and identically distributed (IID). We show how one can build optimal multi-event classifiers from single-event classifiers, and we also show how to construct multi-event classifiers such that they produce optimal single-event classifiers. This is illustrated for a Gaussian example as well as for classification tasks relevant for searches and measurements at the Large Hadron Collider. We extend our discussion to regression tasks by showing how they can be phrased in terms of parametrized classifiers. Empirically, we find that training a single-event (per-instance) classifier is more effective than training a multi-event (per-ensemble) classifier, as least for the cases we studied, and we relate this fact to properties of the loss function gradient in the two cases. While we did not identify a clear benefit from using multi-event classifiers in the collider context, we speculate on the potential value of these methods in cases involving only approximate independence, as relevant for jet substructure studies.

arXiv comment17 pages, 10 figures, 1 table; v2: added footnote about GAN training and added exponential example in appendix; v3: minor updates to match journal version

References in corpus (45)

• Adam: A Method for Stochastic Optimization
• Auto-Encoding Variational Bayes
• PYTHIA 6.4 Physics and Manual
• PYTHIA 6.2 Physics and Manual
• TensorFlow: A system for large-scale machine learning
• The anti-k_t jet clustering algorithm
• A Brief Introduction to PYTHIA 8.1
• FastJet user manual
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• Dispelling the N^3 myth for the Kt jet-finder
• Variational Inference with Normalizing Flows
• Identifying Boosted Objects with N-subjettiness
• Identification of b-quark jets with the CMS experiment
• Deep Learning and its Application to LHC Physics
• Maximizing Boosted Top Identification by Minimizing N-subjettiness
• Jet Substructure at the Large Hadron Collider: A Review of Recent Advances in Theory and Machine Learning
• Anomaly Detection with Density Estimation
• Classification without labels: Learning from mixed samples in high energy physics
• Energy Flow Networks: Deep Sets for Particle Jets
• Parameterized Machine Learning for High-Energy Physics
• Gaining (Mutual) Information about Quark/Gluon Discrimination
• Learning New Physics from a Machine
• OmniFold: A Method to Simultaneously Unfold All Observables
• Mining gold from implicit models to improve likelihood-free inference
• Constraining Effective Field Theories with Machine Learning
• Simulation Assisted Likelihood-free Anomaly Detection
• A Guide to Constraining Effective Field Theories with Machine Learning
• Machine and Deep Learning Applications in Particle Physics
• Measurements of $b$-jet tagging efficiency with the ATLAS detector using $t\bar{t}$ events at $\sqrt{s}=13$ TeV
• The Lund Jet Plane
• Approximating Likelihood Ratios with Calibrated Discriminative Classifiers
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• Neural Networks for Full Phase-space Reweighting and Parameter Tuning
• A Living Review of Machine Learning for Particle Physics
• Identifying the nature of the QCD transition in relativistic collision of heavy nuclei with deep learning
• Explainable machine learning of the underlying physics of high-energy particle collisions
• Parameter Estimation using Neural Networks in the Presence of Detector Effects
• Parametrizing the Detector Response with Neural Networks
• Learning Multivariate New Physics
• MCNNTUNES: tuning Shower Monte Carlo generators with machine learning
• Resonance Searches with Machine Learned Likelihood Ratios
• Parameter Inference from Event Ensembles and the Top-Quark Mass
• Distinguishing $W^{\prime }$ Signals at Hadron Colliders Using Neural Networks
• Efficiency Parameterization with Neural Networks
• Automated Discovery of Jet Substructure Analyses

Cited by in corpus (18)

• CaloFlow: Fast and Accurate Generation of Calorimeter Showers with Normalizing Flows
• Anomaly detection search for new resonances decaying into a Higgs boson and a generic new particle $X$ in hadronic final states using $\sqrt{s}=13$ TeV $pp$ collisions with the ATLAS detector
• An equation-of-state-meter for CBM using PointNet
• Learning Likelihood Ratios with Neural Network Classifiers
• Data-Directed Search for New Physics based on Symmetries of the SM
• Learning Uncertainties the Frequentist Way: Calibration and Correlation in High Energy Physics
• A method for approximating optimal statistical significances with machine-learned likelihoods
• Resonant Anomaly Detection with Multiple Reference Datasets
• Bias and Priors in Machine Learning Calibrations for High Energy Physics
• Towards a method to anticipate dark matter signals with deep learning at the LHC
• Reconstructing partonic kinematics at colliders with Machine Learning
• Neural Conditional Reweighting
• Simulation-based Anomaly Detection for Multileptons at the LHC
• Moment Unfolding
• BitHEP -- The Limits of Low-Precision ML in HEP
• Latent Space Refinement for Deep Generative Models
• Finding excesses in model parameter space
• Applying Machine Learning Techniques To Intermediate-Length Cascade Decays