skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: MLPF: efficient machine-learned particle-flow reconstruction using graph neural networks

Journal Article · · European Physical Journal. C, Particles and Fields
 [1];  [2];  [3];  [4];  [3]
  1. National Inst. of Chemical Physics and Biophysics (NICPB), Tallinn (Estonia); California Institute of Technology (CalTech), Pasadena, CA (United States)
  2. Univ. of California, San Diego, La Jolla, CA (United States)
  3. California Institute of Technology (CalTech), Pasadena, CA (United States)
  4. European Organization for Nuclear Research (CERN), Geneva (Switzerland)
+ Show Author Affiliations

In general-purpose particle detectors, the particle-flow algorithm may be used to reconstruct a comprehensive particle-level view of the event by combining information from the calorimeters and the trackers, significantly improving the detector resolution for jets and the missing transverse momentum. In view of the planned high-luminosity upgrade of the CERN Large Hadron Collider (LHC), it is necessary to revisit existing reconstruction algorithms and ensure that both the physics and computational performance are sufficient in an environment with many simultaneous proton–proton interactions (pileup). Machine learning may offer a prospect for computationally efficient event reconstruction that is well-suited to heterogeneous computing platforms, while significantly improving the reconstruction quality over rule-based algorithms for granular detectors. We introduce MLPF, a novel, end-to-end trainable, machine-learned particle-flow algorithm based on parallelizable, computationally efficient, and scalable graph neural network optimized using a multi-task objective on simulated events. We report the physics and computational performance of the MLPF algorithm on a Monte Carlo dataset of top quark–antiquark pairs produced in proton–proton collisions in conditions similar to those expected for the high-luminosity LHC. The MLPF algorithm improves the physics response with respect to a rule-based benchmark algorithm and demonstrates computationally scalable particle-flow reconstruction in a high-pileup environment.

Research Organization:
California Institute of Technology (CalTech), Pasadena, CA (United States); Univ. of California, San Diego, CA (United States); Fermi National Accelerator Laboratory (FNAL), Batavia, IL (United States)
Sponsoring Organization:
USDOE Office of Science (SC), High Energy Physics (HEP); National Science Foundation (NSF); Estonian Research Council; European Research Council (ERC); European Union’s Horizon 2020
Grant/Contract Number:
SC0011925; SC0019227; SC0021187; SC0021396; AC02-07CH11359; 1624356; MOBTP187; 772369; CNS-1730158; ACI-1540112; ACI-1541349; OAC-1826967
OSTI ID:
1851571
Journal Information:
European Physical Journal. C, Particles and Fields, Vol. 81, Issue 5; ISSN 1434-6044
Publisher:
SpringerCopyright Statement
Country of Publication:
United States
Language:
English

References (49)

FPGAs-as-a-Service Toolkit (FaaST) conference November 2020
Metric Space of Collider Events journal July 2019
Similarity hashing for charged particle tracking conference December 2019
Learning representations of irregular particle-detector geometry with distance-weighted graph networks journal July 2019
Jet flavour classification using DeepJet journal December 2020
Electromagnetic showers beyond shower shapes
  • de Oliveira, Luke; Nachman, Benjamin; Paganini, Michela
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 951 https://doi.org/10.1016/j.nima.2019.162879
journal January 2020
Fast inference of deep neural networks in FPGAs for particle physics journal July 2018
Energy flow networks: deep sets for particle jets journal January 2019
JEDI-net: a jet identification algorithm based on interaction networks journal January 2020
PYTHIA 6.4 physics and manual journal May 2006
GPU coprocessors as a service for deep learning inference in high energy physics journal April 2021
Interaction networks for the identification of boosted H → b b ¯ decays journal July 2020
Simulated particle-level events of ttbar and QCD with PU200 using Pythia8+Delphes3 for machine learned particle flow (MLPF) dataset January 2021
ATLAS b-jet identification performance and efficiency measurement with $$t{\bar{t}}$$tt¯ events in pp collisions at $$\sqrt{s}=13$$s=13 TeV journal November 2019
How to GAN away detector effects journal January 2020
Measurement of the diffractive structure function journal January 1998
Physics requirements for the FCC-hh calorimeter system journal January 2019
Calorimetry with deep learning: particle simulation and reconstruction for collider physics journal July 2020
An analysis of the charged and neutral energy flow in e+e− hadronic annihilation at 34 GeV, and a determination of the QCD effective coupling constant journal July 1982
Dynamic Graph CNN for Learning on Point Clouds journal October 2019
Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics journal January 2021
Measurement of the diffractive cross section in deep inelastic scattering using ZEUS 1994 data journal January 1999
Performance of the DELPHI detector
  • Abreu, P.; Adam, W.; Adye, T.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 378, Issue 1-2 https://doi.org/10.1016/0168-9002(96)00463-9
journal August 1996
Graph neural networks in particle physics journal January 2021
ABCNet: an attention-based method for particle tagging journal June 2020
A brief introduction to PYTHIA 8.1 journal June 2008
How to GAN LHC events journal January 2019
Study of jet Energy Resolution at cdf journal October 2001
DELPHES 3: a modular framework for fast simulation of a generic collider experiment journal February 2014
Jet tagging via particle clouds journal March 2020
A Search for Supersymmetric Higgs Bosons in the Di-tau Decay Mode in Proton - Anti-proton Collisions at 1.8 TeV report January 2003
Supervised jet clustering with graph neural networks for Lorentz boosted bosons journal October 2020
FPGA-Accelerated Machine Learning Inference as a Service for Particle Physics Computing journal October 2019
FCC-ee: The Lepton Collider: Future Circular Collider Conceptual Design Report Volume 2 journal June 2019
Measurement of <mml:math altimg="si1.gif" overflow="scroll" xmlns:xocs="http://www.elsevier.com/xml/xocs/dtd" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://www.elsevier.com/xml/ja/dtd" xmlns:ja="http://www.elsevier.com/xml/ja/dtd" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:tb="http://www.elsevier.com/xml/common/table/dtd" xmlns:sb="http://www.elsevier.com/xml/common/struct-bib/dtd" xmlns:ce="http://www.elsevier.com/xml/common/dtd" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:cals="http://www.elsevier.com/xml/common/cals/dtd"><mml:mi>σ</mml:mi><mml:mo stretchy="false">(</mml:mo><mml:mi>p</mml:mi><mml:mover accent="true"><mml:mi>p</mml:mi><mml:mo>¯</mml:mo></mml:mover><mml:mo>→</mml:mo><mml:mi>Z</mml:mi><mml:mo>+</mml:mo><mml:mi>X</mml:mi><mml:mo stretchy="false">)</mml:mo><mml:mi mathvariant="normal">Br</mml:mi><mml:mo stretchy="false">(</mml:mo><mml:mi>Z</mml:mi><mml:mo>→</mml:mo><mml:msup><mml:mi>τ</mml:mi><mml:mo>+</mml:mo></mml:msup><mml:msup><mml:mi>τ</mml:mi><mml:mo>−</mml:mo></mml:msup><mml:mo stretchy="false">)</mml:mo></mml:math> at <mml:math altimg="si2.gif" overflow="scroll" xmlns:xocs="http://www.elsevier.com/xml/xocs/dtd" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://www.elsevier.com/xml/ja/dtd" xmlns:ja="http://www.elsevier.com/xml/ja/dtd" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:tb="http://www.elsevier.com/xml/common/table/dtd" xmlns:sb="http://www.elsevier.com/xml/common/struct-bib/dtd" xmlns:ce="http://www.elsevier.com/xml/common/dtd" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:cals="http://www.elsevier.com/xml/common/cals/dtd"><mml:msqrt><mml:mi>s</mml:mi></mml:msqrt><mml:mo>=</mml:mo><mml:mn>1.96</mml:mn><mml:mtext> TeV</mml:mtext></mml:math> journal January 2009
Identification of heavy-flavour jets with the CMS detector in pp collisions at 13 TeV journal May 2018
Identification of boosted Higgs bosons decaying into b-quark pairs with the ATLAS detector at 13 $$\text {TeV}$$TeV journal October 2019
Measurement of σ ( p p ¯ → Z ) · B ( Z → τ τ ) in p p ¯ collisions at s = 1.96     TeV journal May 2007
FCC-hh: The Hadron Collider: Future Circular Collider Conceptual Design Report Volume 3 journal July 2019
HepSim: A Repository with Predictions for High-Energy Physics Experiments journal January 2015
Jet reconstruction and performance using particle flow with the ATLAS Detector journal July 2017
Challenges of Particle Flow reconstruction in the CMS High-Granularity Calorimeter at the High-Luminosity LHC journal November 2017
Pileup mitigation at the Large Hadron Collider with graph neural networks journal July 2019
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation conference July 2017
Performance of the ALEPH detector at LEP
  • Buskulic, D.; Casper, D.; De Bonis, I.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 360, Issue 3 https://doi.org/10.1016/0168-9002(95)00138-7
journal June 1995
Particle-flow reconstruction and global event description with the CMS detector journal October 2017
Object condensation: one-stage grid-free multi-object reconstruction in physics detectors, graph, and image data journal September 2020
Graph Neural Networks for Particle Tracking and Reconstruction text January 2020
Simulated particle-level dataset of ttbar with PU200 using Pythia8+Delphes3 for machine learned particle flow (MLPF) dataset January 2021

Cited By (1)

Semi-supervised graph neural networks for pileup noise removal journal January 2023

Similar Records

Particle-based fast jet simulation at the LHC with variational autoencoders
Journal Article · Wed Jul 13 00:00:00 EDT 2022 · Machine Learning: Science and Technology · OSTI ID:1851571
+6 more
Lorentz group equivariant autoencoders
Journal Article · Fri Jun 09 00:00:00 EDT 2023 · European Physical Journal. C, Particles and Fields (Online) · OSTI ID:1851571
+1 more
The Dark Machines Anomaly Score Challenge: Benchmark Data and Model Independent Event Classification for the Large Hadron Collider
Journal Article · Fri Jan 28 00:00:00 EST 2022 · SciPost Physics · OSTI ID:1851571
+36 more

Related Subjects

71 CLASSICAL AND QUANTUM MECHANICS
GENERAL PHYSICS