×

Weakly supervised classification in high energy physics. (English) Zbl 1380.81439

Summary: As machine learning algorithms become increasingly sophisticated to exploit subtle features of the data, they often become more dependent on simulations. This paper presents a new approach called weakly supervised classification in which class proportions are the only input into the machine learning algorithm. Using one of the most challenging binary classification tasks in high energy physics – quark versus gluon tagging – we show that weakly supervised classification can match the performance of fully supervised algorithms. Furthermore, by design, the new algorithm is insensitive to any mis-modeling of discriminating features in the data by the simulation. Weakly supervised classification is a general procedure that can be applied to a wide variety of learning problems to boost performance and robustness when detailed simulations are not reliable or not available.

MSC:

81V05 Strong interaction, including quantum chromodynamics
68T05 Learning and adaptive systems in artificial intelligence
PDFBibTeX XMLCite
Full Text: DOI arXiv

References:

[1] ATLAS collaboration, Performance of b-jet identification in the ATLAS experiment, 2016 JINST11 P04008 [arXiv:1512.01094] [INSPIRE].
[2] CMS collaboration, Identification of b-quark jets with the CMS experiment, 2013 JINST8 P04013 [arXiv:1211.4462] [INSPIRE].
[3] ATLAS collaboration, Light-quark and gluon jet discrimination in pp collisions at \[\sqrt{s}=7 \sqrt{s}=7\] TeV with the ATLAS detector, Eur. Phys. J.C 74 (2014) 3023 [arXiv:1405.6583] [INSPIRE].
[4] CMS collaboration, Performance of quark/gluon discrimination in 8 TeV pp data, CMS-PAS-JME-13-002 (2013) · Zbl 1042.68650
[5] ATLAS collaboration, Identification of boosted, hadronically decaying W bosons and comparisons with ATLAS data taken at \[\sqrt{s}=8 \sqrt{s}=8\] TeV, Eur. Phys. J.C 76 (2016) 154 [arXiv:1510.05821] [INSPIRE]. · Zbl 1393.81007
[6] CMS collaboration, Identification techniques for highly boosted W bosons that decay into hadrons, JHEP12 (2014) 017 [arXiv:1410.4227] [INSPIRE].
[7] ATLAS collaboration, Identification of high transverse momentum top quarks in pp collisions at \[\sqrt{s}=8 \sqrt{s}=8\] TeV with the ATLAS detector, JHEP06 (2016) 093 [arXiv:1603.03127] [INSPIRE].
[8] CMS collaboration, Boosted top jet tagging at CMS, CMS-PAS-JME-13-007 (2013).
[9] T.G. Dietterich, R.H. Lathrop and T. Lozano-Pérez, Solving the multiple instance problem with axis-parallel rectangles, Artif. Intell.89 (1997) 31. · Zbl 1042.68650 · doi:10.1016/S0004-3702(96)00034-3
[10] J. Amores, Multiple instance classification: Review, taxonomy and comparative study, Artif. Intell.201 (2013) 81. · Zbl 1334.68176 · doi:10.1016/j.artint.2013.06.003
[11] D. Kotzias, M. Denil, N. de Freitas and P. Smyth, From group to individual labels using deep features, in the proceedings of the 21thACM SIGKDD international conference on knowledge discovery and data mining (KDD15), August 10-13, Sydney, Australia (2015).
[12] G. Patrini, R. Nock, P. Rivera and T. Caetano, (Almost) No label no cry, in Advances in Neural Information Processing Systems 27, Z. Ghahramani et al. eds., Curran Associates Inc., U.S.A. (2014).
[13] J. Gallicchio and M.D. Schwartz, Quark and gluon tagging at the LHC, Phys. Rev. Lett.107 (2011) 172001 [arXiv:1106.3076] [INSPIRE]. · doi:10.1103/PhysRevLett.107.172001
[14] J.R. Andersen et al., Les Houches 2015: Physics at TeV Colliders Standard Model Working Group Report, in the proceedings of the 9thLes Houches Workshop on Physics at TeV Colliders (PhysTeV 2015), June 1-19, Les Houches, France (2016), arXiv:1605.04692 [INSPIRE].
[15] D.P. Kingma and J. Ba, Adam: A method for stochastic optimization, arXiv:1412.6980.
[16] F. Chollet, Keras, https://github.com/fchollet/keras (2015). · Zbl 1334.68176
[17] CMS collaboration, V tagging observables and correlations, CMS-PAS-JME-14-002 (2014).
[18] ATLAS collaboration, Search for high-mass diboson resonances with boson-tagged jets in proton-proton collisions at \[\sqrt{s}=8 \sqrt{s}=8\] TeV with the ATLAS detector, JHEP12 (2015) 055 [arXiv:1506.00962] [INSPIRE].
[19] CMS collaboration, Search for the standard model Higgs boson produced through vector boson fusion and decaying tobb¯\[ b\overline{b} \], Phys. Rev.D 92 (2015) 032008 [arXiv:1506.01010] [INSPIRE]. · Zbl 1042.68650
[20] CMS collaboration, Measurement of electroweak production of two jets in association with a Z boson in proton-proton collisions at \[\sqrt{s}=8 \sqrt{s}=8\] TeV, Eur. Phys. J.C 75 (2015) 66 [arXiv:1410.3153] [INSPIRE]. · Zbl 1290.81155
[21] ATLAS collaboration, Search for the Standard Model Higgs boson produced by vector-boson fusion and decaying to bottom quarks in \[\sqrt{s}=8 \sqrt{s}=8\] TeV pp collisions with the ATLAS detector, JHEP11 (2016) 112 [arXiv:1606.02181] [INSPIRE].
[22] B. Bhattacherjee, S. Mukhopadhyay, M.M. Nojiri, Y. Sakaki and B.R. Webber, Quark-gluon discrimination in the search for gluino pair production at the LHC, JHEP01 (2017) 044 [arXiv:1609.08781] [INSPIRE]. · doi:10.1007/JHEP01(2017)044
[23] J. Rojo et al., The PDF4LHC report on PDFs and LHC data: Results from Run I and preparation for Run II, J. Phys.G 42 (2015) 103103 [arXiv:1507.00556] [INSPIRE]. · doi:10.1088/0954-3899/42/10/103103
[24] J. Alwall et al., The automated computation of tree-level and next-to-leading order differential cross sections and their matching to parton shower simulations, JHEP07 (2014) 079 [arXiv:1405.0301] [INSPIRE]. · doi:10.1007/JHEP07(2014)079
[25] S. Alioli, P. Nason, C. Oleari and E. Re, A general framework for implementing NLO calculations in shower Monte Carlo programs: the POWHEG BOX, JHEP06 (2010) 043 [arXiv:1002.2581] [INSPIRE]. · Zbl 1290.81155 · doi:10.1007/JHEP06(2010)043
[26] T. Sjöstrand, S. Mrenna and P.Z. Skands, PYTHIA 6.4 physics and manual, JHEP05 (2006) 026 [hep-ph/0603175] [INSPIRE]. · Zbl 1368.81015
[27] M. Bahr et al., HERWIG++ physics and manual, Eur. Phys. J.C 58 (2008) 639 [arXiv:0803.0883] [INSPIRE]. · doi:10.1140/epjc/s10052-008-0798-9
[28] T. Gleisberg et al., Event generation with SHERPA 1.1, JHEP02 (2009) 007 [arXiv:0811.4622] [INSPIRE].
[29] B. Andersson, G. Gustafson, G. Ingelman and T. Sjöstrand, Parton fragmentation and string dynamics, Phys. Rept.97 (1983) 31 [INSPIRE]. · doi:10.1016/0370-1573(83)90080-7
[30] B.R. Webber, A QCD model for jet fragmentation including soft gluon interference, Nucl. Phys.B 238 (1984) 492 [INSPIRE]. · doi:10.1016/0550-3213(84)90333-X
[31] ALEPH collaboration, D. Buskulic et al., Quark and gluon jet properties in symmetric three jet events, Phys. Lett.B 384 (1996) 353 [INSPIRE].
[32] P.T. Komiske, E.M. Metodiev and M.D. Schwartz, Deep learning in color: towards automated quark/gluon jet discrimination, JHEP01 (2017) 110 [arXiv:1612.01551] [INSPIRE]. · Zbl 1373.81388 · doi:10.1007/JHEP01(2017)110
[33] G. Dissertori, I.G. Knowles and M. Schmelling, High energy experiments and theory, Clarendon Press, Oxford U.K. (2003). · Zbl 1071.81001
[34] T. Sjöstrand, S. Mrenna and P.Z. Skands, A brief introduction to PYTHIA 8.1, Comput. Phys. Commun.178 (2008) 852 [arXiv:0710.3820] [INSPIRE]. · Zbl 1196.81038
[35] M. Cacciari, G.P. Salam and G. Soyez, The anti-ktjet clustering algorithm, JHEP04 (2008) 063 [arXiv:0802.1189] [INSPIRE]. · Zbl 1369.81100 · doi:10.1088/1126-6708/2008/04/063
[36] M. Cacciari, G.P. Salam and G. Soyez, FastJet user manual, Eur. Phys. J.C 72 (2012) 1896 [arXiv:1111.6097] [INSPIRE]. · Zbl 1393.81007 · doi:10.1140/epjc/s10052-012-1896-2
[37] Particle Data Group collaboration, K.A. Olive et al., Review of particle physics, Chin. Phys.C 38 (2014) 090001 [INSPIRE]. · Zbl 1373.81388
[38] A.J. Larkoski, J. Thaler and W.J. Waalewijn, Gaining (mutual) information about quark/gluon discrimination, JHEP11 (2014) 129 [arXiv:1408.3122] [INSPIRE]. · doi:10.1007/JHEP11(2014)129
[39] ATLAS collaboration, Measurement of the charged-particle multiplicity inside jets from \[\sqrt{s}=8 \sqrt{s}=8\] TeV pp collisions with the ATLAS detector, Eur. Phys. J.C 76 (2016) 322 [arXiv:1602.00988] [INSPIRE].
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.