MLPF: efficient machine-learned particle-flow reconstruction using graph neural networks
Abstract
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.
- Authors:
-
- National Inst. of Chemical Physics and Biophysics (NICPB), Tallinn (Estonia); California Institute of Technology (CalTech), Pasadena, CA (United States)
- Univ. of California, San Diego, La Jolla, CA (United States)
- California Institute of Technology (CalTech), Pasadena, CA (United States)
- European Organization for Nuclear Research (CERN), Geneva (Switzerland)
- Publication Date:
- Research Org.:
- 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 Org.:
- 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
- OSTI Identifier:
- 1851571
- Grant/Contract Number:
- SC0011925; SC0019227; SC0021187; SC0021396; AC02-07CH11359; 1624356; MOBTP187; 772369; CNS-1730158; ACI-1540112; ACI-1541349; OAC-1826967
- Resource Type:
- Accepted Manuscript
- Journal Name:
- European Physical Journal. C, Particles and Fields
- Additional Journal Information:
- Journal Volume: 81; Journal Issue: 5; Journal ID: ISSN 1434-6044
- Publisher:
- Springer
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSICS
Citation Formats
Pata, Joosep, Duarte, Javier, Vlimant, Jean-Roch, Pierini, Maurizio, and Spiropulu, Maria. MLPF: efficient machine-learned particle-flow reconstruction using graph neural networks. United States: N. p., 2021.
Web. doi:10.1140/epjc/s10052-021-09158-w.
Pata, Joosep, Duarte, Javier, Vlimant, Jean-Roch, Pierini, Maurizio, & Spiropulu, Maria. MLPF: efficient machine-learned particle-flow reconstruction using graph neural networks. United States. https://doi.org/10.1140/epjc/s10052-021-09158-w
Pata, Joosep, Duarte, Javier, Vlimant, Jean-Roch, Pierini, Maurizio, and Spiropulu, Maria. Sun .
"MLPF: efficient machine-learned particle-flow reconstruction using graph neural networks". United States. https://doi.org/10.1140/epjc/s10052-021-09158-w. https://www.osti.gov/servlets/purl/1851571.
@article{osti_1851571,
title = {MLPF: efficient machine-learned particle-flow reconstruction using graph neural networks},
author = {Pata, Joosep and Duarte, Javier and Vlimant, Jean-Roch and Pierini, Maurizio and Spiropulu, Maria},
abstractNote = {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.},
doi = {10.1140/epjc/s10052-021-09158-w},
journal = {European Physical Journal. C, Particles and Fields},
number = 5,
volume = 81,
place = {United States},
year = {Sun May 02 00:00:00 EDT 2021},
month = {Sun May 02 00:00:00 EDT 2021}
}
Works referenced in this record:
FPGAs-as-a-Service Toolkit (FaaST)
conference, November 2020
- Rankin, Dylan; Krupa, Jeffrey; Harris, Philip
- 2020 IEEE/ACM International Workshop on Heterogeneous High-performance Reconfigurable Computing (H2RC)
Metric Space of Collider Events
journal, July 2019
- Komiske, Patrick T.; Metodiev, Eric M.; Thaler, Jesse
- Physical Review Letters, Vol. 123, Issue 4
Similarity hashing for charged particle tracking
conference, December 2019
- Amrouche, Sabrina; Golling, Tobias; Kiehn, Moritz
- 2019 IEEE International Conference on Big Data (Big Data)
Learning representations of irregular particle-detector geometry with distance-weighted graph networks
journal, July 2019
- Qasim, Shah Rukh; Kieseler, Jan; Iiyama, Yutaro
- The European Physical Journal C, Vol. 79, Issue 7
Jet flavour classification using DeepJet
journal, December 2020
- Bols, E.; Kieseler, J.; Verzetti, M.
- Journal of Instrumentation, Vol. 15, Issue 12
Electromagnetic showers beyond shower shapes
journal, January 2020
- de Oliveira, Luke; Nachman, Benjamin; Paganini, Michela
- Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 951
Fast inference of deep neural networks in FPGAs for particle physics
journal, July 2018
- Duarte, J.; Han, S.; Harris, P.
- Journal of Instrumentation, Vol. 13, Issue 07
Energy flow networks: deep sets for particle jets
journal, January 2019
- Komiske, Patrick T.; Metodiev, Eric M.; Thaler, Jesse
- Journal of High Energy Physics, Vol. 2019, Issue 1
JEDI-net: a jet identification algorithm based on interaction networks
journal, January 2020
- Moreno, Eric A.; Cerri, Olmo; Duarte, Javier M.
- The European Physical Journal C, Vol. 80, Issue 1
PYTHIA 6.4 physics and manual
journal, May 2006
- Sjöstrand, Torbjörn; Mrenna, Stephen; Skands, Peter
- Journal of High Energy Physics, Vol. 2006, Issue 05
GPU coprocessors as a service for deep learning inference in high energy physics
journal, April 2021
- Krupa, Jeffrey; Lin, Kelvin; Acosta Flechas, Maria
- Machine Learning: Science and Technology, Vol. 2, Issue 3
Interaction networks for the identification of boosted decays
journal, July 2020
- Moreno, Eric A.; Nguyen, Thong Q.; Vlimant, Jean-Roch
- Physical Review D, Vol. 102, Issue 1
Simulated particle-level events of ttbar and QCD with PU200 using Pythia8+Delphes3 for machine learned particle flow (MLPF)
dataset, January 2021
- Pata, Joosep; Duarte, Javier Mauricio; Vlimant, Jean-Roch
- Zenodo
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
- Aad, G.; Abbott, B.; Abbott, D. C.
- The European Physical Journal C, Vol. 79, Issue 11
How to GAN away detector effects
journal, January 2020
- Bellagente, Marco; Butter, Anja; Kasieczka, Gregor
- SciPost Physics, Vol. 8, Issue 4
Measurement of the diffractive structure function
journal, January 1998
- Breitweg et al., J.
- The European Physical Journal C, Vol. 1, Issue 1
Physics requirements for the FCC-hh calorimeter system
journal, January 2019
- Selvaggi, Michele
- Journal of Physics: Conference Series, Vol. 1162
Calorimetry with deep learning: particle simulation and reconstruction for collider physics
journal, July 2020
- Belayneh, Dawit; Carminati, Federico; Farbin, Amir
- The European Physical Journal C, Vol. 80, Issue 7
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
- Behrend, H. J.; Chen, Ch.; Fenner, H.
- Physics Letters B, Vol. 113, Issue 5
Dynamic Graph CNN for Learning on Point Clouds
journal, October 2019
- Wang, Yue; Sun, Yongbin; Liu, Ziwei
- ACM Transactions on Graphics, Vol. 38, Issue 5
Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics
journal, January 2021
- Iiyama, Yutaro; Cerminara, Gianluca; Gupta, Abhijay
- Frontiers in Big Data, Vol. 3
Measurement of the diffractive cross section in deep inelastic scattering using ZEUS 1994 data
journal, January 1999
- Breitweg et al., J.
- The European Physical Journal C, Vol. 6, Issue 1
Performance of the DELPHI detector
journal, August 1996
- 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
Graph neural networks in particle physics
journal, January 2021
- Shlomi, Jonathan; Battaglia, Peter; Vlimant, Jean-Roch
- Machine Learning: Science and Technology, Vol. 2, Issue 2
ABCNet: an attention-based method for particle tagging
journal, June 2020
- Mikuni, V.; Canelli, F.
- The European Physical Journal Plus, Vol. 135, Issue 6
A brief introduction to PYTHIA 8.1
journal, June 2008
- Sjöstrand, Torbjörn; Mrenna, Stephen; Skands, Peter
- Computer Physics Communications, Vol. 178, Issue 11
How to GAN LHC events
journal, January 2019
- Butter, Anja; Plehn, Tilman; Winterhalder, Ramon
- SciPost Physics, Vol. 7, Issue 6
Study of jet Energy Resolution at cdf
journal, October 2001
- Bocci, A.; Kuhlmann, S.; Lami, S.
- International Journal of Modern Physics A, Vol. 16, Issue supp01a
DELPHES 3: a modular framework for fast simulation of a generic collider experiment
journal, February 2014
- de Favereau, J.; Delaere, C.; Demin, P.
- Journal of High Energy Physics, Vol. 2014, Issue 2
Jet tagging via particle clouds
journal, March 2020
- Qu, Huilin; Gouskos, Loukas
- Physical Review D, Vol. 101, Issue 5
A Search for Supersymmetric Higgs Bosons in the Di-tau Decay Mode in Proton - Anti-proton Collisions at 1.8 TeV
report, January 2003
- Connolly, Amy Lynn
Supervised jet clustering with graph neural networks for Lorentz boosted bosons
journal, October 2020
- Ju, Xiangyang; Nachman, Benjamin
- Physical Review D, Vol. 102, Issue 7
FPGA-Accelerated Machine Learning Inference as a Service for Particle Physics Computing
journal, October 2019
- Duarte, Javier; Harris, Philip; Hauck, Scott
- Computing and Software for Big Science, Vol. 3, Issue 1
FCC-ee: The Lepton Collider: Future Circular Collider Conceptual Design Report Volume 2
journal, June 2019
- Abada, A.; Abbrescia, M.; AbdusSalam, S. S.
- The European Physical Journal Special Topics, Vol. 228, Issue 2
Identification of heavy-flavour jets with the CMS detector in pp collisions at 13 TeV
journal, May 2018
- Sirunyan, A. M.; Tumasyan, A.; Adam, W.
- Journal of Instrumentation, Vol. 13, Issue 05
Identification of boosted Higgs bosons decaying into b-quark pairs with the ATLAS detector at 13 $$\text {TeV}$$TeV
journal, October 2019
- Aad, G.; Abbott, B.; Abbott, D. C.
- The European Physical Journal C, Vol. 79, Issue 10
Measurement of in collisions at
journal, May 2007
- Abulencia, A.; Adelman, J.; Affolder, T.
- Physical Review D, Vol. 75, Issue 9
FCC-hh: The Hadron Collider: Future Circular Collider Conceptual Design Report Volume 3
journal, July 2019
- Abada, A.; Abbrescia, M.; AbdusSalam, S. S.
- The European Physical Journal Special Topics, Vol. 228, Issue 4
HepSim: A Repository with Predictions for High-Energy Physics Experiments
journal, January 2015
- Chekanov, S. V.
- Advances in High Energy Physics, Vol. 2015
Jet reconstruction and performance using particle flow with the ATLAS Detector
journal, July 2017
- Aaboud, M.; Aad, G.; Abbott, B.
- The European Physical Journal C, Vol. 77, Issue 7
Challenges of Particle Flow reconstruction in the CMS High-Granularity Calorimeter at the High-Luminosity LHC
journal, November 2017
- Chlebana, Frank
- Journal of Physics: Conference Series, Vol. 928
Pileup mitigation at the Large Hadron Collider with graph neural networks
journal, July 2019
- Arjona Martínez, J.; Cerri, O.; Spiropulu, M.
- The European Physical Journal Plus, Vol. 134, Issue 7
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
conference, July 2017
- Charles, R. Qi; Su, Hao; Kaichun, Mo
- 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Performance of the ALEPH detector at LEP
journal, June 1995
- 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
Particle-flow reconstruction and global event description with the CMS detector
journal, October 2017
- Sirunyan, A. M.; Tumasyan, A.; Adam, W.
- Journal of Instrumentation, Vol. 12, Issue 10
Object condensation: one-stage grid-free multi-object reconstruction in physics detectors, graph, and image data
journal, September 2020
- Kieseler, Jan
- The European Physical Journal C, Vol. 80, Issue 9
Object condensation: one-stage grid-free multi-object reconstruction in physics detectors, graph, and image data
journal, September 2020
- Kieseler, Jan
- The European Physical Journal C, Vol. 80, Issue 9
Graph Neural Networks for Particle Tracking and Reconstruction
text, January 2020
- Duarte, Javier; Vlimant, Jean-Roch
- arXiv
Simulated particle-level dataset of ttbar with PU200 using Pythia8+Delphes3 for machine learned particle flow (MLPF)
dataset, January 2021
- Pata, Joosep; Duarte, Javier Mauricio; Vlimant, Jean-Roch
- Zenodo
Works referencing / citing this record:
Semi-supervised graph neural networks for pileup noise removal
journal, January 2023
- Li, Tianchun; Liu, Shikun; Feng, Yongbin
- The European Physical Journal C, Vol. 83, Issue 1