Boosted top tagging and its interpretation using Shapley values
arXiv:2212.11606 · doi:10.1140/epjp/s13360-024-05910-9
Abstract
Top tagging has emerged as a fast-evolving subject due to the top quark's significant role in probing physics beyond the standard model. For the reconstruction of top jets, machine learning models have shown a substantial improvement in the classification performance compared to the previous methods. In this work, we build top taggers using -Subjettiness ratios and several Energy Correlation observables as input features to train the eXtreme Gradient BOOSTed decision tree (XGBOOST). The study finds that tighter parton-level matching lead to more accurate tagging. However, in real experimental data, where the parton level data are unknown, this matching cannot be done. We train the XGBOOST models without performing this matching and show that this difference impacts the taggers' effectiveness. Additionally, we test the tagger under different simulation conditions, including changes in center-of-mass energy, parton distribution functions (PDFs), and pileup effects, demonstrating its robustness with performance deviations of less than 1%. Furthermore, we use the SHapley Additive exPlanation (SHAP) framework to calculate the importance of the features of the trained models. It helps us to estimate how much each feature of the data contributed to the model's prediction and what regions are of more importance for each input variable. Finally, we combine all the tagger variables to form a hybrid tagger and interpret the results using the Shapley values.
55 pages, 32 figures, 17 tables
References in corpus (49)
- XGBoost: A Scalable Tree Boosting System
- The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
- A Brief Introduction to PYTHIA 8.1
- FastJet user manual
- DELPHES 3, A modular framework for fast simulation of a generic collider experiment
- MadGraph 5 : Going Beyond
- Herwig++ Physics and Manual
- LHAPDF6: parton density access in the LHC precision era
- Dispelling the N^3 myth for the Kt jet-finder
- Herwig 7.0 / Herwig++ 3.0 Release Note
- Electroweak symmetry breaking from dimensional deconstruction
- Better Jet Clustering Algorithms
- The Littlest Higgs
- Identifying Boosted Objects with N-subjettiness
- Top-tagging: A Method for Identifying Boosted Hadronic Tops
- ParticleNet: Jet Tagging via Particle Clouds
- Energy Correlation Functions for Jet Substructure
- Pileup Per Particle Identification
- Recombination Algorithms and Jet Substructure: Pruning as a Tool for Heavy Particle Searches
- Fat Jets for a Light Higgs
- Stop Reconstruction with Tagged Tops
- Jet-Images -- Deep Learning Edition
- Deep-learning Top Taggers or The End of QCD?
- The Machine Learning Landscape of Top Taggers
- Power Counting to Better Jet Observables
- Deep-learned Top Tagging with a Lorentz Layer
- New Angles on Energy Correlation Functions
- What is the Machine Learning?
- Pulling Out All the Tops with Computer Vision and Deep Learning
- A likelihood-based reconstruction algorithm for top-quark pairs and the KLFitter framework
- Measurement of the top quark mass in the lepton+jets channel from TeV ATLAS data and combination with previous results
- Search for anomalous t t-bar production in the highly-boosted all-hadronic final state
- How to Improve Top Tagging
- Associated jet and subjet rates in light-quark and gluon jet discrimination
- SPANet: Generalized Permutationless Set Assignment for Particle Physics using Symmetry Preserving Attention
- Measurement of lepton differential distributions and the top quark mass in production in collisions at TeV with the ATLAS detector
- Building a Better Boosted Top Tagger
- Boosted decision trees in the era of new physics: a smuon analysis case study
- Interpretable machine learning in Physics
- Neural Network-based Top Tagger with Two-Point Energy Correlations and Geometry of Soft Emissions
- Reports of My Demise Are Greatly Exaggerated: -subjettiness Taggers Take On Jet Images
- Jet-Parton Assignment in ttH Events using Deep Learning
- CapsNets Continuing the Convolutional Quest
- Leveraging universality of jet taggers through transfer learning
- Boosted top quark tagging and polarization measurement using machine learning
- Measurement of the Top Quark Mass Using the Invariant Mass of Lepton Pairs in Soft Muon b-tagged Events
- Interpretable Machine Learning Methods Applied to Jet Background Subtraction in Heavy Ion Collisions
- Automatic detection of boosted Higgs boson and top quark jets in an event image
- Investigating the Violation of Charge Parity Symmetry Through Top Quark Chromo-Electric Dipole Moments by Using Machine Learning Techniques