Interpretable Deep Learning for Two-Prong Jet Classification with Jet Spectra

Abstract

Classification of jets with deep learning has gained significant attention in recent times. However, the performance of deep neural networks is often achieved at the cost of interpretability. Here we propose an interpretable network trained on the jet spectrum $S_2(R)$ which is a two-point correlation function of the jet constituents. The spectrum can be derived from a functional Taylor series of an arbitrary jet classifier function of energy flows. An interpretable network can be obtained by truncating the series. The intermediate feature of the network is an infrared and collinear safe C-correlator which allows us to estimate the importance of a $S_2(R)$ deposit at an angular scale R in the classification. The performance of the architecture is comparable to that of a convolutional neural network (CNN) trained on jet images, although the number of inputs and complexity of architecture is significantly simpler than the CNN classifier. We consider two examples: one is the classification of two-prong jets which differ in color charge of the mother particle, and the other is a comparison between Pythia 8 and Herwig 7 generated jets.

arXiv comment32 pages, 21 figures, published in JHEP

References in corpus (80)

• Adam: A Method for Stochastic Optimization
• 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
• An Introduction to PYTHIA 8.2
• FastJet user manual
• Geometric deep learning: going beyond Euclidean data
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• FeynRules 2.0 - A complete toolbox for tree-level phenomenology
• Herwig++ Physics and Manual
• Parton distributions with LHC data
• Dispelling the N^3 myth for the Kt jet-finder
• UFO - The Universal FeynRules Output
• Herwig 7.0 / Herwig++ 3.0 Release Note
• Jet substructure as a new Higgs search channel at the LHC
• Tuning PYTHIA 8.1: the Monash 2013 Tune
• Identifying Boosted Objects with N-subjettiness
• Soft Drop
• Model-Agnostic Interpretability of Machine Learning
• Jet Trimming
• Towards an understanding of jet substructure
• Top-tagging: A Method for Identifying Boosted Hadronic Tops
• ParticleNet: Jet Tagging via Particle Clouds
• Towards Robust Interpretability with Self-Explaining Neural Networks
• Energy Correlation Functions for Jet Substructure
• Jet substructure as a new Higgs search channel at the LHC
• Recombination Algorithms and Jet Substructure: Pruning as a Tool for Heavy Particle Searches
• Fat Jets for a Light Higgs
• Techniques for improved heavy particle searches with jet substructure
• Jet Substructure at the Large Hadron Collider: Experimental Review
• Classification without labels: Learning from mixed samples in high energy physics
• Stop Reconstruction with Tagged Tops
• Deep learning in color: towards automated quark/gluon jet discrimination
• Jet-Images -- Deep Learning Edition
• Deep-learning Top Taggers or The End of QCD?
• Strategies to Identify Boosted Tops
• Searching for New Physics with Deep Autoencoders
• Anomaly Detection for Resonant New Physics with Machine Learning
• Energy Flow Networks: Deep Sets for Particle Jets
• Colour reconnections in Herwig++
• Power Counting to Better Jet Observables
• Quark and Gluon Tagging at the LHC
• Learning New Physics from a Machine
• Extending the Bump Hunt with Machine Learning
• Deep-learned Top Tagging with a Lorentz Layer
• Seeing in Color: Jet Superstructure
• New Angles on Energy Correlation Functions
• Jet-Images: Computer Vision Inspired Techniques for Jet Tagging
• QCD or What?
• Constraining Effective Field Theories with Machine Learning
• Novelty Detection Meets Collider Physics
• A Guide to Constraining Effective Field Theories with Machine Learning
• Energy flow polynomials: A complete linear basis for jet substructure
• Finding top quarks with shower deconstruction
• Finding physics signals with shower deconstruction
• Guiding New Physics Searches with Unsupervised Learning
• The Lund Jet Plane
• Decorrelated Jet Substructure Tagging using Adversarial Neural Networks
• Variational Autoencoders for New Physics Mining at the Large Hadron Collider
• QCD-Aware Recursive Neural Networks for Jet Physics
• Pileup Mitigation with Machine Learning (PUMML)
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• Boosting $H\to b\bar{b}$ with Machine Learning
• Weakly Supervised Classification in High Energy Physics
• Learning to Pivot with Adversarial Networks
• Recursive Neural Networks in Quark/Gluon Tagging
• Measuring Multijet Structure of Hadronic Energy Flow Or What IS A Jet?
• Finding physics signals with event deconstruction
• (Machine) Learning to Do More with Less
• Quark-Gluon Tagging: Machine Learning vs Detector
• Associated jet and subjet rates in light-quark and gluon jet discrimination
• Automating the Construction of Jet Observables with Machine Learning
• Automated next-to-leading order predictions for new physics at the LHC: the case of colored scalar pair production
• Infrared Safety of a Neural-Net Top Tagging Algorithm
• Opening the black box of neural nets: case studies in stop/top discrimination
• Approaching robust EFT limits for CP-violation in the Higgs sector
• Spectral Analysis of Jet Substructure with Neural Networks: Boosted Higgs Case
• Deep learning for the R-parity violating supersymmetry searches at the LHC
• Telescoping Jets: Multiple Event Interpretations with Multiple R's
• A Framework for Finding Anomalous Objects at the LHC
• QCD Jets and Parton Showers

Cited by in corpus (30)

• An Efficient Lorentz Equivariant Graph Neural Network for Jet Tagging
• Searches for the BSM scenarios at the LHC using decision tree based machine learning algorithms: A comparative study and review of Random Forest, Adaboost, XGboost and LightGBM frameworks
• Exploring the Space of Jets with CMS Open Data
• ABCDisCo: Automating the ABCD Method with Machine Learning
• A guide for deploying Deep Learning in LHC searches: How to achieve optimality and account for uncertainty
• Supervised Jet Clustering with Graph Neural Networks for Lorentz Boosted Bosons
• Equivariant Energy Flow Networks for Jet Tagging
• Symmetries, Safety, and Self-Supervision
• The Boosted Higgs Jet Reconstruction via Graph Neural Network
• Energy-weighted Message Passing: an infra-red and collinear safe graph neural network algorithm
• Constraints on the trilinear and quartic Higgs couplings from triple Higgs production at the LHC and beyond
• Applications of deep learning to relativistic hydrodynamics
• CapsNets Continuing the Convolutional Quest
• Study of energy deposition patterns in hadron calorimeter for prompt and displaced jets using convolutional neural network
• Leveraging universality of jet taggers through transfer learning
• A Detailed Study of Interpretability of Deep Neural Network based Top Taggers
• Quantum Generative Adversarial Networks For Anomaly Detection In High Energy Physics
• Maximum performance of strange-jet tagging at hadron colliders
• Jet Discrimination with Quantum Complete Graph Neural Network
• Comparing Point Cloud Strategies for Collider Event Classification
• Deep Learning Jet Substructure from Two-Particle Correlation
• Jet Classification Using High-Level Features from Anatomy of Top Jets
• Role of polarizations and spin-spin correlations of $W^{\prime }s$ in $e^{-}{e}^{+}\to W^{-}{W}^{+}$ at $\sqrt{s}=250$ GeV to probe anomalous $W^{-}{W}^{+}Z/\gamma$ couplings
• Interplay of Traditional Methods and Machine Learning Algorithms for Tagging Boosted Objects
• Explainable AI for ML jet taggers using expert variables and layerwise relevance propagation
• Jet tagging in the Lund plane with graph networks
• Riemannian Data preprocessing in Machine Learning to focus on QCD color structure
• Identifying the Quantum Properties of Hadronic Resonances using Machine Learning
• Tagging the Higgs boson decay to bottom quarks with colour-sensitive observables and the Lund jet plane
• Reweighting and Analysing Event Generator Systematics by Neural Networks on High-Level Features