Learning to increase matching efficiency in identifying additional b-jets in the $$\text {t}\bar{\text {t}}\text {b}\bar{\text {b}}$$ t

The tt̄H(bb̄) process is an essential channel to reveal the Higgs properties but has an irreducible background from the tt̄bb̄ process, which produces a top quark pair in association with a b quark pair. Therefore, understanding the tt̄bb̄ process is crucial for improving the sensitivity of a search for the tt̄H(bb̄) process. To this end, when measuring the differential cross-section of the tt̄bb̄ process, we need to distinguish the b-jets originated from top quark decays, and additional b-jets originated from gluon splitting. Since there are no simple identification rules, we adopt deep learning methods to learn from data to identify the additional b-jets from the tt̄bb̄ events. Specifically, by exploiting the special structure of the tt̄bb̄ event data, we propose several loss functions that can be minimized to directly increase the matching efficiency, the accuracy of identifying additional b-jets. We discuss the difference between our method and another deep learning-based approach based on binary classification [1] using synthetic data. We then verify that additional b-jets can be identified more accurately by increasing matching efficiency directly rather than the binary classification accuracy, using simulated tt̄bb̄ event data in the lepton+jets channel from pp collision at √ s = 13 TeV.

[1]  M. Cacciari,et al.  The anti-$k_t$ jet clustering algorithm , 2008, 0802.1189.

[2]  L. F. Chaparro Sierra,et al.  Observation of tt[over ¯]H Production. , 2018, Physical review letters.

[3]  P. Baldi,et al.  Jet flavor classification in high-energy physics with deep neural networks , 2016, 1607.08633.

[4]  Eli Upfal,et al.  Machine Learning in High Energy Physics Community White Paper , 2018, Journal of Physics: Conference Series.

[5]  R. Frederix,et al.  The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations , 2014, 1405.0301.

[6]  J. Favereau,et al.  DELPHES 3: a modular framework for fast simulation of a generic collider experiment , 2013, Journal of High Energy Physics.

[7]  C. Collaboration,et al.  Particle-flow reconstruction and global event description with the CMS detector , 2017, 1706.04965.

[8]  C. Collaboration,et al.  Measurement of tt production with additional jet activity, including b quark jets, in the dilepton decay channel using pp collisions at √ s = 8 TeV , 2015 .

[9]  Kyunghyun Cho,et al.  QCD-aware recursive neural networks for jet physics , 2017, Journal of High Energy Physics.

[10]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[11]  S. M. Etesami,et al.  Measurement of the ttbb production cross section in the all-jet final state in pp collisions at √ s = 13 TeV The CMS Collaboration , 2019 .

[12]  The Cms Collaboration.  Observation of a new boson at a mass of 125 GeV with the CMS experiment at the LHC , 2012, 1207.7235.

[13]  R. Frederix,et al.  Automatic spin-entangled decays of heavy resonances in Monte Carlo simulations , 2012, 1212.3460.

[14]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[15]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[16]  J. Cogan,et al.  Jet-images: computer vision inspired techniques for jet tagging , 2014, 1407.5675.

[17]  Peter Skands,et al.  An introduction to PYTHIA 8.2 , 2014, Comput. Phys. Commun..

[18]  Leandro Giordano Almeida,et al.  Playing tag with ANN: boosted top identification with pattern recognition , 2015, 1501.05968.

[19]  Kazuhiro Terao,et al.  Machine learning at the energy and intensity frontiers of particle physics , 2018, Nature.

[20]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[21]  Atlas Collaboration.  Observation of Higgs boson production in association with a top quark pair at the LHC with the ATLAS detector , 2018, 1806.00425.

[22]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[23]  D. Whiteson,et al.  Deep Learning and Its Application to LHC Physics , 2018, Annual Review of Nuclear and Particle Science.

[24]  D. Bourilkov.  Machine and deep learning applications in particle physics , 2019, International Journal of Modern Physics A.

[25]  Luke de Oliveira,et al.  Jet-images — deep learning edition , 2015, Journal of High Energy Physics.

[26]  C. Collaboration,et al.  Identification of heavy-flavour jets with the CMS detector in pp collisions at 13 TeV , 2017, Journal of Instrumentation.

[27]  Cern İşbirliği.  Measurements of tt¯ cross sections in association with b jets and inclusive jets and their ratio using dilepton final states in pp collisions at s=13TeV , 2017 .

[28]  B. Nachman,et al.  Jet substructure at the Large Hadron Collider: A review of recent advances in theory and machine learning , 2017, Physics Reports.