Pulling out all the tops with computer vision and deep learning

Macaluso, Sebastian; Shih, David

doi:10.1007/JHEP10(2018)121

Title: Pulling out all the tops with computer vision and deep learning

Journal Article · Thu Oct 18 00:00:00 EDT 2018 · Journal of High Energy Physics (Online)

DOI:https://doi.org/10.1007/JHEP10(2018)121· OSTI ID:1483579

Macaluso, Sebastian ^[1]; Shih, David ^[1]

Rutgers Univ., Piscataway, NJ (United States)

We apply computer vision with deep learning — in the form of a convolutional neural network (CNN) — to build a highly effective boosted top tagger. Previous work (the “DeepTop” tagger of Kasieczka et al) has shown that a CNN-based top tagger can achieve comparable performance to state-of-the-art conventional top taggers based on high-level inputs. Here, we introduce a number of improvements to the DeepTop tagger, including architecture, training, image preprocessing, sample size and color pixels. Our final CNN top tagger outperforms BDTs based on high-level inputs by a factor of ~ 2–3 or more in background rejection, over a wide range of tagging efficiencies and fiducial jet selections. As reference points, we achieve a QCD background rejection factor of 500 (60) at 50%top tagging efficiency for fully-merged (non-merged) top jets with pT in the 800–900 GeV (350–450 GeV) range. Our CNN can also be straightforwardly extended to the classification of other types of jets, and the lessons learned here may be useful to others designing their own deep NNs for LHC applications.

View Accepted Manuscript (DOE)

Cite

Export

Save

Research Organization:: Rutgers Univ., Piscataway, NJ (United States)

Sponsoring Organization:: USDOE

Grant/Contract Number:: SC0010008

OSTI ID:: 1483579

Journal Information:: Journal of High Energy Physics (Online), Vol. 2018, Issue 10; ISSN 1029-8479

Publisher:: Springer BerlinCopyright Statement

Country of Publication:: United States

Language:: English

Citation Metrics:

Cited by: 77 works

Citation information provided by
Web of Science

References (57)

Deep learning LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey Nature, Vol. 521, Issue 7553 https://doi.org/10.1038/nature14539	journal	May 2015
Deep-learned Top Tagging with a Lorentz Layer Butter, Anja; Kasieczka, Gregor; Plehn, Tilman SciPost Physics, Vol. 5, Issue 3 https://doi.org/10.21468/SciPostPhys.5.3.028	journal	January 2018
Jet substructure classification in high-energy physics with deep neural networks Baldi, Pierre; Bauer, Kevin; Eng, Clara Physical Review D, Vol. 93, Issue 9 https://doi.org/10.1103/PhysRevD.93.094034	journal	May 2016
Finding physics signals with shower deconstruction Soper, Davison E.; Spannowsky, Michael Physical Review D, Vol. 84, Issue 7 https://doi.org/10.1103/PhysRevD.84.074002	journal	October 2011
Boosted objects: a probe of beyond the standard model physics Abdesselam, A.; Belyaev, A.; Kuutmann, E. Bergeaas The European Physical Journal C, Vol. 71, Issue 6 https://doi.org/10.1140/epjc/s10052-011-1661-y	journal	June 2011
Resonance searches with an updated top tagger Kasieczka, Gregor; Plehn, Tilman; Schell, Torben ETH Zurich https://doi.org/10.3929/ethz-b-000106218	text	January 2015
Deep learning in color: towards automated quark/gluon jet discrimination Komiske, Patrick T.; Metodiev, Eric M.; Schwartz, Matthew D. Journal of High Energy Physics, Vol. 2017, Issue 1 https://doi.org/10.1007/JHEP01(2017)110	journal	January 2017
Deep-learning top taggers or the end of QCD? Kasieczka, Gregor; Plehn, Tilman; Russell, Michael Journal of High Energy Physics, Vol. 2017, Issue 5 https://doi.org/10.1007/JHEP05(2017)006	journal	May 2017
FastJet user manual: (for version 3.0.2) Cacciari, Matteo; Salam, Gavin P.; Soyez, Gregory The European Physical Journal C, Vol. 72, Issue 3 https://doi.org/10.1140/epjc/s10052-012-1896-2	journal	March 2012
Search for W ′  →  tb decays in the hadronic final state using pp collisions at $\sqrt{s} = 13 TeV$ with the ATLAS detector Aaboud, M.; Aad, G.; Abbott, B. Physics Letters B, Vol. 781 https://doi.org/10.1016/j.physletb.2018.03.036	journal	June 2018
Boosted objects and jet substructure at the LHC. Report of BOOST2012, held at IFIC Valencia, 23rd–27th of July 2012 Altheimer, A.; Arce, A.; Asquith, L. The European Physical Journal C, Vol. 74, Issue 3 https://doi.org/10.1140/epjc/s10052-014-2792-8	journal	March 2014
Top tagging Plehn, Tilman; Spannowsky, Michael Journal of Physics G: Nuclear and Particle Physics, Vol. 39, Issue 8 https://doi.org/10.1088/0954-3899/39/8/083001	journal	May 2012
Identifying boosted objects with N-subjettiness Thaler, Jesse; Van Tilburg, Ken Journal of High Energy Physics, Vol. 2011, Issue 3 https://doi.org/10.1007/JHEP03(2011)015	journal	March 2011
Jet Substructure Shelton, J. Proceedings of the 2012 Theoretical Advanced Study Institute in Elementary Particle Physics, Searching for New Physics at Small and Large Scales https://doi.org/10.1142/9789814525220_0007	conference	November 2013
Resonance searches with an updated top tagger Kasieczka, Gregor; Plehn, Tilman; Schell, Torben Journal of High Energy Physics, Vol. 2015, Issue 6 https://doi.org/10.1007/JHEP06(2015)203	journal	June 2015
Show and tell: A neural image caption generator Vinyals, Oriol; Toshev, Alexander; Bengio, Samy 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2015.7298935	conference	June 2015
Accelerating Science with Generative Adversarial Networks: An Application to 3D Particle Showers in Multilayer Calorimeters Paganini, Michela; de Oliveira, Luke; Nachman, Benjamin Physical Review Letters, Vol. 120, Issue 4 https://doi.org/10.1103/PhysRevLett.120.042003	journal	January 2018
Herwig++ physics and manual Bähr, Manuel; Gieseke, Stefan; Gigg, Martyn A. The European Physical Journal C, Vol. 58, Issue 4 https://doi.org/10.1140/epjc/s10052-008-0798-9	journal	November 2008
Learning to classify from impure samples with high-dimensional data Komiske, Patrick T.; Metodiev, Eric M.; Nachman, Benjamin Physical Review D, Vol. 98, Issue 1 https://doi.org/10.1103/PhysRevD.98.011502	journal	July 2018
Towards an understanding of the correlations in jet substructure: Report of BOOST2013, hosted by the University of Arizona, 12th–16th of August 2013 Adams, D.; Arce, A.; Asquith, L. The European Physical Journal C, Vol. 75, Issue 9 https://doi.org/10.1140/epjc/s10052-015-3587-2	journal	September 2015
Jet Substructure as a New Higgs-Search Channel at the Large Hadron Collider Butterworth, Jonathan M.; Davison, Adam R.; Rubin, Mathieu Physical Review Letters, Vol. 100, Issue 24 https://doi.org/10.1103/PhysRevLett.100.242001	journal	June 2008
Deep-learned Top Tagging with a Lorentz Layer Butter, Anja; Kasieczka, Gregor; Plehn, Tilman ETH Zurich https://doi.org/10.3929/ethz-b-000415419	text	January 2018
Learning long-range vision for autonomous off-road driving Hadsell, Raia; Sermanet, Pierre; Ben, Jan Journal of Field Robotics, Vol. 26, Issue 2 https://doi.org/10.1002/rob.20276	journal	February 2009
Soft drop Larkoski, Andrew J.; Marzani, Simone; Soyez, Gregory Journal of High Energy Physics, Vol. 2014, Issue 5 https://doi.org/10.1007/JHEP05(2014)146	journal	May 2014
Jet substructure at the Tevatron and LHC: new results, new tools, new benchmarks Altheimer, A.; Arora, S.; Asquith, L. Journal of Physics G: Nuclear and Particle Physics, Vol. 39, Issue 6 https://doi.org/10.1088/0954-3899/39/6/063001	journal	May 2012
Top Tagging: A Method for Identifying Boosted Hadronically Decaying Top Quarks Kaplan, David E.; Rehermann, Keith; Schwartz, Matthew D. Physical Review Letters, Vol. 101, Issue 14 https://doi.org/10.1103/PhysRevLett.101.142001	journal	October 2008
CaloGAN: Simulating 3D high energy particle showers in multilayer electromagnetic calorimeters with generative adversarial networks Paganini, Michela; de Oliveira, Luke; Nachman, Benjamin Physical Review D, Vol. 97, Issue 1 https://doi.org/10.1103/PhysRevD.97.014021	journal	January 2018
Measurements of $t \bar{t}$ differential cross-sections of highly boosted top quarks decaying to all-hadronic final states in $p p$ collisions at $\sqrt{s} = 13 TeV$ using the ATLAS detector Aaboud, M.; Aad, G.; Abbott, B. Physical Review D, Vol. 98, Issue 1 https://doi.org/10.1103/PhysRevD.98.012003	journal	July 2018
Classification without labels: learning from mixed samples in high energy physics Metodiev, Eric M.; Nachman, Benjamin; Thaler, Jesse Journal of High Energy Physics, Vol. 2017, Issue 10 https://doi.org/10.1007/JHEP10(2017)174	journal	October 2017
Towards jetography Salam, Gavin P. The European Physical Journal C, Vol. 67, Issue 3-4 https://doi.org/10.1140/epjc/s10052-010-1314-6	journal	May 2010
Jet-images — deep learning edition de Oliveira, Luke; Kagan, Michael; Mackey, Lester Journal of High Energy Physics, Vol. 2016, Issue 7 https://doi.org/10.1007/JHEP07(2016)069	journal	July 2016
Weakly supervised classification in high energy physics Dery, Lucio Mwinmaarong; Nachman, Benjamin; Rubbo, Francesco Journal of High Energy Physics, Vol. 2017, Issue 5 https://doi.org/10.1007/JHEP05(2017)145	journal	May 2017
Finding top quarks with shower deconstruction Soper, Davison E.; Spannowsky, Michael Physical Review D, Vol. 87, Issue 5 https://doi.org/10.1103/PhysRevD.87.054012	journal	March 2013
Jet-images: computer vision inspired techniques for jet tagging Cogan, Josh; Kagan, Michael; Strauss, Emanuel Journal of High Energy Physics, Vol. 2015, Issue 2 https://doi.org/10.1007/JHEP02(2015)118	journal	February 2015
Deep Residual Learning for Image Recognition He, Kaiming; Zhang, Xiangyu; Ren, Shaoqing 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2016.90	conference	June 2016
Herwig 7.0/Herwig++ 3.0 release note Bellm, Johannes; Gieseke, Stefan; Grellscheid, David The European Physical Journal C, Vol. 76, Issue 4 https://doi.org/10.1140/epjc/s10052-016-4018-8	journal	April 2016
Fat Jets for a Light Higgs Boson Plehn, Tilman; Salam, Gavin P.; Spannowsky, Michael Physical Review Letters, Vol. 104, Issue 11 https://doi.org/10.1103/PhysRevLett.104.111801	journal	March 2010
An introduction to PYTHIA 8.2 Sjöstrand, Torbjörn; Ask, Stefan; Christiansen, Jesper R. Computer Physics Communications, Vol. 191 https://doi.org/10.1016/j.cpc.2015.01.024	journal	June 2015
Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis de Oliveira, Luke; Paganini, Michela; Nachman, Benjamin Computing and Software for Big Science, Vol. 1, Issue 1 https://doi.org/10.1007/s41781-017-0004-6	journal	September 2017
Playing tag with ANN: boosted top identification with pattern recognition Almeida, Leandro G.; Backović, Mihailo; Cliche, Mathieu Journal of High Energy Physics, Vol. 2015, Issue 7 https://doi.org/10.1007/JHEP07(2015)086	journal	July 2015
Fast R-CNN Girshick, Ross 2015 IEEE International Conference on Computer Vision (ICCV) https://doi.org/10.1109/ICCV.2015.169	conference	December 2015
Geant4—a simulation toolkit Agostinelli, S.; Allison, J.; Amako, K. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 506, Issue 3 https://doi.org/10.1016/S0168-9002(03)01368-8	journal	July 2003
Phenomenological and theoretical developments in jet physics at the LHC Cacciari, Matteo International Journal of Modern Physics A, Vol. 30, Issue 31 https://doi.org/10.1142/S0217751X1546001X	journal	November 2015
DeepFace: Closing the Gap to Human-Level Performance in Face Verification Taigman, Yaniv; Yang, Ming; Ranzato, Marc'Aurelio 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2014.220	conference	June 2014
DELPHES 3: a modular framework for fast simulation of a generic collider experiment de Favereau, J.; Delaere, C.; Demin, P. Journal of High Energy Physics, Vol. 2014, Issue 2 https://doi.org/10.1007/JHEP02(2014)057	journal	February 2014
(Machine) learning to do more with less Cohen, Timothy; Freytsis, Marat; Ostdiek, Bryan Journal of High Energy Physics, Vol. 2018, Issue 2 https://doi.org/10.1007/JHEP02(2018)034	journal	February 2018
Deep-learning top taggers or the end of QCD Kasieczka, Gregor; Plehn, Tilman; Russell, Michael ETH Zurich https://doi.org/10.3929/ethz-b-000191242	text	January 2017
Deep Learning Ganatra, Sanket Zenodo https://doi.org/10.5281/zenodo.1189359	text	January 2018
Herwig++ Physics and Manual Bahr, M.; Gieseke, S.; Gigg, M. A. arXiv https://doi.org/10.48550/arxiv.0803.0883	text	January 2008
Stop Reconstruction with Tagged Tops Plehn, Tilman; Spannowsky, Michael; Takeuchi, Michihisa arXiv https://doi.org/10.48550/arxiv.1006.2833	text	January 2010
Jet-Images: Computer Vision Inspired Techniques for Jet Tagging Cogan, Josh; Kagan, Michael; Strauss, Emanuel arXiv https://doi.org/10.48550/arxiv.1407.5675	text	January 2014
Show and Tell: A Neural Image Caption Generator Vinyals, Oriol; Toshev, Alexander; Bengio, Samy arXiv https://doi.org/10.48550/arxiv.1411.4555	preprint	January 2014
Deep Residual Learning for Image Recognition He, Kaiming; Zhang, Xiangyu; Ren, Shaoqing arXiv https://doi.org/10.48550/arxiv.1512.03385	preprint	January 2015
Deep learning in color: towards automated quark/gluon jet discrimination Komiske, Patrick T.; Metodiev, Eric M.; Schwartz, Matthew D. arXiv https://doi.org/10.48550/arxiv.1612.01551	text	January 2016
Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis de Oliveira, Luke; Paganini, Michela; Nachman, Benjamin arXiv https://doi.org/10.48550/arxiv.1701.05927	text	January 2017
Deep-learning Top Taggers or The End of QCD? Kasieczka, Gregor; Plehn, Tilman; Russell, Michael arXiv https://doi.org/10.48550/arxiv.1701.08784	text	January 2017
Weakly Supervised Classification in High Energy Physics Dery, Lucio Mwinmaarong; Nachman, Benjamin; Rubbo, Francesco arXiv https://doi.org/10.48550/arxiv.1702.00414	text	January 2017

Cited By (17)

Identification of heavy, energetic, hadronically decaying particles using machine-learning techniques Sirunyan, Albert M.; Tumasyan, Armen; Adam, Wolfgang Deutsches Elektronen-Synchrotron, DESY, Hamburg https://doi.org/10.3204/pubdb-2020-02644	text	January 2020
Six top messages of new physics at the LHC Han, Huayong; Huang, Li; Ma, Teng Journal of High Energy Physics, Vol. 2019, Issue 10 https://doi.org/10.1007/jhep10(2019)008	journal	October 2019
JEDI-net: a jet identification algorithm based on interaction networks Moreno, Eric A.; Cerri, Olmo; Duarte, Javier M. The European Physical Journal C, Vol. 80, Issue 1 https://doi.org/10.1140/epjc/s10052-020-7608-4	journal	January 2020
Performance of top-quark and $$\varvec{W}$$ W -boson tagging with ATLAS in Run 2 of the LHC Aaboud, M.; Aad, G.; Abbott, B. The European Physical Journal C, Vol. 79, Issue 5 https://doi.org/10.1140/epjc/s10052-019-6847-8	journal	April 2019
Automating the construction of jet observables with machine learning Datta, Kaustuv; Larkoski, Andrew; Nachman, Benjamin ETH Zurich https://doi.org/10.3929/ethz-b-000380014	text	January 2019
Topology Classification with Deep Learning to Improve Real-Time Event Selection at the LHC Nguyen, T. Q.; Weitekamp, D.; Anderson, D. Computing and Software for Big Science, Vol. 3, Issue 1 https://doi.org/10.1007/s41781-019-0028-1	journal	August 2019
The Machine Learning landscape of top taggers Kasieczka, Gregor; Plehn, Tilman; Butter, Anja RWTH Aachen University https://doi.org/10.18154/rwth-2019-07568	text	January 2019
Identification of heavy, energetic, hadronically decaying particles using machine-learning techniques Collaboration, Cms; Sirunyan, Albert M.; Tumasyan, Armen R. ETH Zurich https://doi.org/10.3929/ethz-b-000429003	text	January 2020
Infrared Safety of a Neural-Net Top Tagging Algorithm Choi, Suyong; Lee, Seung J.; Perelstein, Maxim arXiv https://doi.org/10.48550/arxiv.1806.01263	text	January 2018
Topology classification with deep learning to improve real-time event selection at the LHC Nguyen, Thong Q.; Weitekamp, Daniel; Anderson, Dustin arXiv https://doi.org/10.48550/arxiv.1807.00083	text	January 2018
Reports of My Demise Are Greatly Exaggerated: $N$-subjettiness Taggers Take On Jet Images Moore, Liam; Nordström, Karl; Varma, Sreedevi arXiv https://doi.org/10.48550/arxiv.1807.04769	text	January 2018
Reweighting a parton shower using a neural network: the final-state case Bothmann, Enrico; Del Debbio, Luigi arXiv https://doi.org/10.48550/arxiv.1808.07802	text	January 2018
Energy Flow Networks: Deep Sets for Particle Jets Komiske, Patrick T.; Metodiev, Eric M.; Thaler, Jesse arXiv https://doi.org/10.48550/arxiv.1810.05165	text	January 2018
Quark-Gluon Tagging: Machine Learning vs Detector Kasieczka, Gregor; Kiefer, Nicholas; Plehn, Tilman arXiv https://doi.org/10.48550/arxiv.1812.09223	text	January 2018
Automating the Construction of Jet Observables with Machine Learning Datta, Kaustuv; Larkoski, Andrew; Nachman, Benjamin arXiv https://doi.org/10.48550/arxiv.1902.07180	text	January 2019
CapsNets Continuing the Convolutional Quest Diefenbacher, Sascha; Frost, Hermann; Kasieczka, Gregor arXiv https://doi.org/10.48550/arxiv.1906.11265	text	January 2019
JEDI-net: a jet identification algorithm based on interaction networks Moreno, Eric A.; Cerri, Olmo; Duarte, Javier M. arXiv https://doi.org/10.48550/arxiv.1908.05318	text	January 2019

Similar Records

Morphology for jet classification

Journal Article · Wed Jan 05 00:00:00 EST 2022 · Physical Review D · OSTI ID:1483579

Lim, Sung Hak; Nojiri, Mihoko M.

A new method to distinguish hadronically decaying boosted Z bosons from W bosons using the ATLAS detector

Journal Article · Thu Apr 28 00:00:00 EDT 2016 · European Physical Journal. C, Particles and Fields · OSTI ID:1483579

Aad, G.; Abbott, B.; Abdallah, J.; +2855 more

A deep neural network to search for new long-lived particles decaying to jets

Journal Article · Thu Aug 20 00:00:00 EDT 2020 · Machine Learning: Science and Technology · OSTI ID:1483579

Sirunyan, A. M.; Tumasyan, A.; Adam, W.; +2 more

Related Subjects

97 MATHEMATICS AND COMPUTING
Jets

Title: Pulling out all the tops with computer vision and deep learning

Citation Formats

References (57)

Cited By (17)

Similar Records

Related Subjects