Jet tagging made easy

Abstract

We develop taggers for multi-pronged jets that are simple functions of jet substructure (so-called `subjettiness') variables. These taggers can be approximately decorrelated from the jet mass in a quite simple way. Specifically, we use a Logistic Regression Design (LoRD) which, even being one of the simplest machine learning classifiers, shows a performance which surpasses that of simple variables used by the ATLAS and CMS Collaborations and is not far from more complex models based on neural networks. Contrary to the latter, our method allows for an easy implementation of tagging tasks by providing a simple and interpretable analytical formula with already optimised parameters.

arXiv commentLaTeX 28 pages, four extra appendices. Final version in EPJC. Source code available at https://github.com/bzaldivarm/LoRD

References in corpus (37)

• The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
• 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
• Successive Combination Jet Algorithm For Hadron Collisions
• Jet substructure as a new Higgs search channel at the LHC
• Identifying Boosted Objects with N-subjettiness
• Top-tagging: A Method for Identifying Boosted Hadronic Tops
• Energy Correlation Functions for Jet Substructure
• Maximizing Boosted Top Identification by Minimizing N-subjettiness
• Pileup Per Particle Identification
• Fat Jets for a Light Higgs
• Anomaly Detection with Density Estimation
• Stop Reconstruction with Tagged Tops
• Searching for New Physics with Deep Autoencoders
• Strategies to Identify Boosted Tops
• Substructure of high-p_T Jets at the LHC
• Extending the Bump Hunt with Machine Learning
• New Angles on Energy Correlation Functions
• QCD or What?
• A generic anti-QCD jet tagger
• Simulation Assisted Likelihood-free Anomaly Detection
• Energy flow polynomials: A complete linear basis for jet substructure
• Variational Autoencoders for New Physics Mining at the Large Hadron Collider
• Thinking outside the ROCs: Designing Decorrelated Taggers (DDT) for jet substructure
• How Much Information is in a Jet?
• Jet Substructure Without Trees
• Novel Jet Observables from Machine Learning
• Recursive Soft Drop
• Automating the Construction of Jet Observables with Machine Learning
• Mass Agnostic Jet Taggers
• Multiphoton signals of a (96 GeV?) stealth boson
• Reports of My Demise Are Greatly Exaggerated: $N$-subjettiness Taggers Take On Jet Images
• Stealth multiboson signals
• The minimal stealth boson: models and benchmarks
• Running bumps from stealth bosons

Cited by in corpus (9)

• Simulation-Assisted Decorrelation for Resonant Anomaly Detection
• Mapping Machine-Learned Physics into a Human-Readable Space
• Mass Unspecific Supervised Tagging (MUST) for boosted jets
• Higgs boson tagging with the Lund jet plane
• Anomaly detection from mass unspecific jet tagging
• Anomaly Awareness
• Dynamic Radius Jet Clustering Algorithm
• Exotic diboson $Z^{\prime }$ decays in the U$\mu \nu$SSM
• Pulling the Higgs and Top needles from the jet stack with Feature Extended Supervised Tagging