CaltechAUTHORS
Published January 28, 2022 | Version Published + Accepted Version
Journal Article Open

The Dark Machines Anomaly Score Challenge: Benchmark Data and Model Independent Event Classification for the Large Hadron Collider

Creators

  • 1. ROR icon European Organization for Nuclear Research
  • 2. ROR icon University of Oxford
  • 3. ROR icon Queen Mary University of London
  • 4. ROR icon The Ohio State University
  • 5. ROR icon National Institute for Subatomic Physics
  • 6. ROR icon International School for Advanced Studies
  • 7. ROR icon National Institute for Nuclear Physics
  • 8. ROR icon Lund University
  • 9. ROR icon University of California, San Diego
  • 10. ROR icon The University of Texas at Arlington
  • 11. ROR icon Google (United States)
  • 12. ROR icon University of Glasgow
  • 13. ROR icon Worcester Polytechnic Institute
  • 14. ROR icon Konkuk University
  • 15. ROR icon University of Adelaide
  • 16. ROR icon Institute for Corpuscular Physics
  • 17. ROR icon Rice University
  • 18. ROR icon RWTH Aachen University
  • 19. ROR icon California Institute of Technology
  • 20. ROR icon Fermilab
  • 21. ROR icon Harvard University
  • 22. ROR icon AI Institute for Artificial Intelligence and Fundamental Interactions
  • 23. ROR icon Kyungpook National University
  • 24. ROR icon National and Kapodistrian University of Athens
  • 25. ROR icon University of Houston
  • 26. ROR icon University College London
  • 27. ROR icon University of Vienna

Abstract

We describe the outcome of a data challenge conducted as part of the Dark Machines Initiative and the Les Houches 2019 workshop on Physics at TeV colliders. The challenged aims at detecting signals of new physics at the LHC using unsupervised machine learning algorithms. First, we propose how an anomaly score could be implemented to define model-independent signal regions in LHC searches. We define and describe a large benchmark dataset, consisting of >1 Billion simulated LHC events corresponding to 10 fb⁻¹ of proton-proton collisions at a center-of-mass energy of 13 TeV. We then review a wide range of anomaly detection and density estimation algorithms, developed in the context of the data challenge, and we measure their performance in a set of realistic analysis environments. We draw a number of useful conclusions that will aid the development of unsupervised new physics searches during the third run of the LHC, and provide our benchmark dataset for future studies at https://www.phenoMLdata.org. Code to reproduce the analysis is provided at https://github.com/bostdiek/DarkMachines-UnsupervisedChallenge.

Additional Information

© 2022 T. Aarrestad et al. This work is licensed under the Creative Commons Attribution 4.0 International License. Published by the SciPost Foundation. Received 21-06-2021. Accepted 23-12-2021. Published 28-01-2022. MvB acknowledges support from the Science and Technology Facilities Council (grant number ST/T000864/1). The work of A.J. is supported by the National Research Foundation of Korea, Grant No. NRF-2019R1A2C1009419. The work of J.M. is supported in part by the Generalitat Valenciana (GV) through the contract APOSTD/2019/165, and by Spanish and European funds under the project PGC2018-094856-B-I00 (MCIU/AEI/FEDER, EU). The work of R.RdA. and J.M. is supported by the Artemisa project, co-funded by the European Union through the 2014-2020 FEDER Operative Programme of Comunitat Valenciana, project IDIFEDER/2018/048. The work of J.H. and B.R. is supported by the Royal Society under grant URF\R1\191524. The work of C.D. is supported by the Swedish Research Council. Research by A.B. is supported by the US Department of Energy under contract DE-SC0011726. The work of B.O. is supported by the U.S. Department of Energy under contract DE-SC0013607. This work is supported by the National Science Foundation under Cooperative Agreement PHY-2019786 (The NSF AI Institute for Artificial Intelligence and Fundamental Interactions, http://iaifi.org/). The work of P.J. was supported by the DIANA-HEP Graduate Fellowship program and the ThinkSwiss Research Scholarship. P.J. M.P., M.T., and K.A.W are supported by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement no 772369. J.M.D. is supported by the DOE, Office of Science, Office of High Energy Physics Early Career Research program under Award No. DE-SC0021187 and by the DOE, Office of Advanced Scientific Computing Research under Award No. DE-SC0021396 (FAIR4HEP). J-R. V. is supported by the U.S. DOE, Office of Science, Office of High Energy Physics under Award No. DE-SC0011925, DE-SC0019227, and DE-AC02-07CH11359. Computations in this paper were run on the FASRC Cannon cluster supported by the FAS Division of Science Research Computing Group at Harvard University. M.W. and A.L. are supported by the Australian Research Council Discovery Project DP180102209 and the ARC Centre of Excellence in Dark Matter Particle Physics (CE200100008). RV acknowledges support from the European Research Council (ERC) under the European Unions Horizon 2020 research and innovation programme (grant agreement No. 788223, PanScales), and from the Science and Technology Facilities Council (STFC) under the grant ST/P000274/1.

Attached Files

Published - SciPostPhys_12_1_043.pdf

Accepted Version - 2105.14027.pdf

Files

2105.14027.pdf

Files (5.8 MB)

Name Size Download all
md5:85f02fbbb2db67a21ea7d2666f1edbe6
3.3 MB Preview Download
md5:7df76d0d32d63bfebfd9c6f10799efe5
2.5 MB Preview Download

Additional details

Identifiers

Eprint ID
116216
Resolver ID
CaltechAUTHORS:20220810-254273000

Related works

Describes
10.21468/SciPostPhys.12.1.043 (DOI)
https://arxiv.org/abs/2105.14027 (URL)
https://www.phenomldata.org/ (URL)
https://github.com/bostdiek/DarkMachines-UnsupervisedChallenge (URL)

Funding

Science and Technology Facilities Council (STFC)
ST/T000864/1
National Research Foundation of Korea
NRF-2019R1A2C1009419
Generalitat Valenciana
APOSTD/2019/165
Ministerio de Ciencia, Innovación y Universidades (MCIU)
PGC2018-094856-B-I00
European Regional Development Fund
IFEDER/2018/048
Royal Society
URF\R1\191524
Swedish Research Council
Department of Energy (DOE)
DE-SC0011726
Department of Energy (DOE)
DE-SC0013607
NSF
PHY-2019786
DIANA Fellows
ThinkSwiss
European Research Council (ERC)
772369
Department of Energy (DOE)
DE-SC0021187
Department of Energy (DOE)
DE-SC0021396
Department of Energy (DOE)
DE-SC0011925
Department of Energy (DOE)
DE-SC0019227
Department of Energy (DOE)
DE-AC02-07CH11359
Harvard University
Australian Research Council
DP180102209
Australian Research Council
CE200100008
European Research Council (ERC)
788223
Science and Technology Facilities Council (STFC)
ST/P000274/1

Dates

Created
2022-08-12
Created from EPrint's datestamp field
Updated
2023-10-23
Created from EPrint's last_modified field