Interpretable deep learning models for the inference and classification of LHC data

Abstract

The Shower Deconstruction methodology is pivotal in distinguishing signal and background jets, leveraging the detailed information from perturbative parton showers. Rooted in the Neyman-Pearson lemma, this method is theoretically designed to differentiate between signal and background processes optimally in high-energy physics experiments. A key challenge, however, arises from the combinatorial growth associated with increasing jet constituents, which hampers its computational feasibility. We address this by demonstrating that the likelihood derived from comparing the most probable signal and background shower histories is equally effective for discrimination as the conventional approach of summing over all potential histories in top quark versus Quantum Chromodynamics (QCD) scenarios. We propose a novel approach by conceptualising the identification of the most probable shower history as a Markov Decision Process (MDP). Utilising a sophisticated modular point-transformer architecture, our method efficiently learns the optimal policy for this task. The developed neural agent excels in constructing the most likely shower history and demonstrates robust generalisation capabilities on unencountered test data. Remarkably, our approach mitigates the complexity inherent in the inference process, achieving a linear scaling relationship with the number of jet constituents. This offers a computationally viable and theoretically sound method for signal-background differentiation, paving the way for more effective data analysis in particle physics.

arXiv commentExtended discussion in section 2.1. Results unchanged. Matches journal version

References in corpus (42)

• The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
• FastJet user manual
• Dispelling the N^3 myth for the Kt jet-finder
• Better Jet Clustering Algorithms
• Automatic spin-entangled decays of heavy resonances in Monte Carlo simulations
• General-purpose event generators for LHC physics
• A Precision Measurement of the Mass of the Top Quark
• Energy Flow Networks: Deep Sets for Particle Jets
• Jet Flavour Classification Using DeepJet
• What is the Machine Learning?
• Adversarially-trained autoencoders for robust unsupervised new physics searches
• Parton Shower Uncertainties in Jet Substructure Analyses with Deep Neural Networks
• Finding top quarks with shower deconstruction
• Finding physics signals with shower deconstruction
• An Efficient Lorentz Equivariant Graph Neural Network for Jet Tagging
• Automation of the matrix element reweighting method
• MLPF: Efficient machine-learned particle-flow reconstruction using graph neural networks
• Finding physics signals with event deconstruction
• A factorisation-aware Matrix element emulator
• Quark-Gluon tagging with Shower Deconstruction: Unearthing dark matter and Higgs couplings
• A guide for deploying Deep Learning in LHC searches: How to achieve optimality and account for uncertainty
• Deep-Learning Jets with Uncertainties and More
• Uncertainty Aware Learning for High Energy Physics
• Mapping Machine-Learned Physics into a Human-Readable Space
• INFERNO: Inference-Aware Neural Optimisation
• Polishing a shiny Higgs with matrix elements
• Two Invertible Networks for the Matrix Element Method
• Energy-weighted Message Passing: an infra-red and collinear safe graph neural network algorithm
• Learning the language of QCD jets with transformers
• Lorentz group equivariant autoencoders
• Combine and Conquer: Event Reconstruction with Bayesian Ensemble Neural Networks
• Unbinned multivariate observables for global SMEFT analyses from machine learning
• Learning Likelihood Ratios with Neural Network Classifiers
• Jet grooming through reinforcement learning
• Improved Constraints on Effective Top Quark Interactions using Edge Convolution Networks
• Matrix Element Regression with Deep Neural Networks -- breaking the CPU barrier
• Deep Learning Jet Image as a Probe of Light Higgsino Dark Matter at the LHC
• On the BSM reach of four top production at the LHC
• Punzi-loss: A non-differentiable metric approximation for sensitivity optimisation in the search for new particles
• Resolving Combinatorial Ambiguities in Dilepton $t\bar{t}$ Event Topologies with Neural Networks
• Phenomenology at the Large Hadron Collider with Deep Learning: the case of vector-like quarks decaying to light jets
• Influence of QCD parton shower in deep learning invisible Higgs through vector boson fusion

Cited by in corpus (4)

• Interplay of Traditional Methods and Machine Learning Algorithms for Tagging Boosted Objects
• Foundations of automatic feature extraction at LHC--point clouds and graphs
• Stable and Interpretable Jet Physics with IRC-Safe Equivariant Feature Extraction
• Tagging ultra-boosted jets at FCC-hh using machine learning techniques