Deep Image Reconstruction for Background Subtraction in Heavy-Ion Collisions

Abstract

Jet reconstruction in an ultra-relativistic heavy-ion collision suffers from a notoriously large, fluctuating thermal background. Traditional background subtraction methods struggle to remove this soft background while preserving the jet's hard substructure. In this Letter, we present DeepSub, the first machine learning-based approach for full-event background subtraction. DeepSub utilizes a model based on Swin Transformer layers to denoise jet images and disentangle hard jets from the heavy-ion background. DeepSub significantly outperforms existing subtraction techniques by reproducing key jet observables such as jet $p_{\mathrm{T}}$ and mass, and substructure observables such as girth and the energy correlation function, at the sub-percent to percent level. As such, DeepSub paves the way for precision heavy-ion measurements in hitherto inaccessible kinematic regimes.

References in corpus (51)

• The anti-k_t jet clustering algorithm
• FastJet user manual
• Dispelling the N^3 myth for the Kt jet-finder
• Suppression of Hadrons with Large Transverse Momentum in Central Au+Au Collisions at $sqrt {s sub {sub {NN}}}_ = 130 GeV
• Observation of a Centrality-Dependent Dijet Asymmetry in Lead-Lead Collisions at sqrt(S(NN))= 2.76 TeV with the ATLAS Detector at the LHC
• Transverse momentum and collision energy dependence of high $p_T$ hadron suppression in Au+Au collisions at ultrarelativistic energies
• Observation and studies of jet quenching in PbPb collisions at nucleon-nucleon center-of-mass energy = 2.76 TeV
• Disappearance of back-to-back high $p_T$ hadron correlations in central Au+Au collisions at $\sqrt{s_{NN}}$ = 200 GeV
• Centrality dependence of high pT hadron suppression in Au+Au collisions at sqrt(sNN)=130 GeV
• Suppression of charged particle production at large transverse momentum in central Pb-Pb collisions at $\sqrt{s_{\mathrm{NN}}}=2.76$ TeV
• Study of high-pT charged particle suppression in PbPb compared to pp collisions at sqrt(sNN)=2.76 TeV
• Suppressed pi^0 Production at Large Transverse Momentum in Central Au+Au Collisions at sqrt(s_NN) = 200 GeV
• Energy Correlation Functions for Jet Substructure
• The theory and phenomenology of perturbative QCD based jet quenching
• Jet quenching in high-energy heavy-ion collisions
• Deep learning in color: towards automated quark/gluon jet discrimination
• Jet-Images -- Deep Learning Edition
• Jet momentum dependence of jet quenching in PbPb collisions at sqrt(sNN)=2.76 TeV
• Charged-particle nuclear modification factors in PbPb and pPb collisions at sqrt(s[NN]) = 5.02 TeV
• Modification of jet shapes in PbPb collisions at sqrt(s[NN]) = 2.76 TeV
• Transverse momentum spectra and nuclear modification factors of charged particles in pp, p-Pb and Pb-Pb collisions at the LHC
• Jet physics in heavy-ion collisions
• Measurement of jet fragmentation in PbPb and pp collisions at sqrt(s[NN]) = 2.76 TeV
• A perturbative framework for jet quenching
• Jet-Images: Computer Vision Inspired Techniques for Jet Tagging
• Measurement of charged-particle spectra in Pb+Pb collisions at $\sqrt{s_{\mathsf{NN}}}=2.76$ TeV with the ATLAS detector at the LHC
• Medium response in JEWEL and its impact on jet shape observables in heavy ion collisions
• Effects of jet-induced medium excitation in $\gamma$-hadron correlation in A+A collisions
• Measurement of charged jet suppression in Pb-Pb collisions at $\sqrt{s_{\mathrm{NN}}}=2.76$ TeV
• Pulling Out All the Tops with Computer Vision and Deep Learning
• Particle-yield modification in jet-like azimuthal di-hadron correlations in Pb-Pb collisions at $\sqrt{s_{\mathrm{NN}}}$ = 2.76 TeV
• Pileup Mitigation with Machine Learning (PUMML)
• Measurement of the splitting function in pp and PbPb collisions at $\sqrt{s_{{}_{\mathrm{NN}}}}=$ 5.02 TeV
• Centrality dependence of the pseudorapidity density distribution for charged particles in Pb-Pb collisions at $\sqrt{s_{\mathrm{NN}}}=5.02$ TeV
• Measurement of event background fluctuations for charged particle jet reconstruction in Pb-Pb collisions at $\sqrt{s_{\mathrm{NN}}}=2.76$ TeV
• Boosting $H\to b\bar{b}$ with Machine Learning
• Particle-level pileup subtraction for jets and jet shapes
• Measurement of jet fragmentation in Pb+Pb and $pp$ collisions at $\sqrt{s_{\mathrm{NN}}}=2.76$ TeV with the ATLAS detector
• Measurement of transverse momentum relative to dijet systems in PbPb and pp collisions at sqrt(s[NN]) = 2.76 TeV
• Pileup and Underlying Event Mitigation with Iterative Constituent Subtraction
• Simulating V+jet processes in heavy ion collisions with JEWEL
• Deep learning jet modifications in heavy-ion collisions
• Measurement of angular and momentum distributions of charged particles within and around jets in Pb+Pb and $pp$ collisions at $\sqrt{s_{\mathrm{NN}}}=5.02$ TeV with the ATLAS detector
• Machine Learning based jet momentum reconstruction in heavy-ion collisions
• Deep learning assisted jet tomography for the study of Mach cones in QGP
• Differential measurements of jet substructure and partonic energy loss in Au$+$Au collisions at $\sqrt{s_{\mathrm{NN}}}=200$ GeV
• Observation of nuclear modification of energy-energy correlators inside jets in heavy ion collisions
• Measurement of jet radial profiles in Pb$-$Pb collisions at $\sqrt{s_{\mathrm{NN}}}$ = 2.76 TeV
• Identifying quenched jets in heavy ion collisions with machine learning
• Jet substructure observables for jet quenching in Quark Gluon Plasma: a Machine Learning driven analysis
• An array-oriented Python interface for FastJet

Cited by in corpus (1)

• Deep learning for jet modification in the presence of the QGP background