A Detailed Study of Interpretability of Deep Neural Network based Top Taggers

Abstract

Recent developments in the methods of explainable AI (XAI) allow researchers to explore the inner workings of deep neural networks (DNNs), revealing crucial information about input-output relationships and realizing how data connects with machine learning models. In this paper we explore interpretability of DNN models designed to identify jets coming from top quark decay in high energy proton-proton collisions at the Large Hadron Collider (LHC). We review a subset of existing top tagger models and explore different quantitative methods to identify which features play the most important roles in identifying the top jets. We also investigate how and why feature importance varies across different XAI metrics, how correlations among features impact their explainability, and how latent space representations encode information as well as correlate with physically meaningful quantities. Our studies uncover some major pitfalls of existing XAI methods and illustrate how they can be overcome to obtain consistent and meaningful interpretation of these models. We additionally illustrate the activity of hidden layers as Neural Activation Pattern (NAP) diagrams and demonstrate how they can be used to understand how DNNs relay information across the layers and how this understanding can help to make such models significantly simpler by allowing effective model reoptimization and hyperparameter tuning. These studies not only facilitate a methodological approach to interpreting models but also unveil new insights about what these models learn. Incorporating these observations into augmented model design, we propose the Particle Flow Interaction Network (PFIN) model and demonstrate how interpretability-inspired model augmentation can improve top tagging performance.

Repository: https://github.com/FAIR4HEP/xAI4toptagger. Some figure cosmetics have been changed. Accepted at Machine Learning: Science and Technology

References in corpus (17)

• An Introduction to PYTHIA 8.2
• Top-tagging: A Method for Identifying Boosted Hadronic Tops
• Top Jets at the LHC
• An Efficient Lorentz Equivariant Graph Neural Network for Jet Tagging
• Template Overlap Method for Massive Jets
• Lorentz Group Equivariant Neural Network for Particle Physics
• Bump Hunting in Latent Space
• Particle Transformer for Jet Tagging
• Lorentz Boost Networks: Autonomous Physics-Inspired Feature Engineering
• Interpretable machine learning in Physics
• Accelerated Charged Particle Tracking with Graph Neural Networks on FPGAs
• Bridging the Gap Between Explainable AI and Uncertainty Quantification to Enhance Trustability
• PELICAN: Permutation Equivariant and Lorentz Invariant or Covariant Aggregator Network for Particle Physics
• Explainable AI for High Energy Physics
• Snowmass 2021 Computational Frontier CompF03 Topical Group Report: Machine Learning
• Do graph neural networks learn traditional jet substructure?
• Invariance-based Multi-Clustering of Latent Space Embeddings for Equivariant Learning

Cited by in corpus (6)

• Constraints on the trilinear and quartic Higgs couplings from triple Higgs production at the LHC and beyond
• FAIR AI Models in High Energy Physics
• Is infrared-collinear safe information all you need for jet classification?
• Interplay of Traditional Methods and Machine Learning Algorithms for Tagging Boosted Objects
• The Physics Behind ML-based Quark-Gluon Taggers
• Graph theory inspired anomaly detection at the LHC