A guide for deploying Deep Learning in LHC searches: How to achieve optimality and account for uncertainty

Deep learning tools can incorporate all of the available information into a search for new particles, thus making the best use of the available data. This paper reviews how to optimally integrate information with deep learning and explicitly describes the corresponding sources of uncertainty. Simple illustrative examples show how these concepts can be applied in practice.

[1]  G. Menardi,et al.  Nonparametric semi-supervised classification with application to signal detection in high energy physics , 2021, Statistical Methods & Applications.

[2]  Gregor Kasieczka,et al.  Robust Jet Classifiers through Distance Correlation. , 2020, Physical review letters.

[3]  S. M. Etesami,et al.  Search for direct top squark pair production in events with one lepton, jets, and missing transverse momentum at 13 TeV with the CMS experiment , 2020, Journal of High Energy Physics.

[4]  O. Amram,et al.  Tag N’ Train: a technique to train improved classifiers on unlabeled data , 2020, Journal of High Energy Physics.

[5]  B. Nachman,et al.  Simulation assisted likelihood-free anomaly detection , 2020, Physical Review D.

[6]  B. Nachman,et al.  Anomaly detection with density estimation , 2020, Physical Review D.

[7]  G. Kasieczka,et al.  DisCo Fever: Robust Networks Through Distance Correlation , 2020, 2001.05310.

[8]  M. Pierini,et al.  Learning multivariate new physics , 2019, The European Physical Journal C.

[9]  M. White,et al.  Does SUSY have friends? A new approach for LHC event analysis , 2019, Journal of High Energy Physics.

[10]  A. Butter,et al.  How to GAN event subtraction , 2019, SciPost Physics Core.

[11]  Maurizio Pierini,et al.  Particle Generative Adversarial Networks for full-event simulation at the LHC and their application to pileup description , 2019, Journal of Physics: Conference Series.

[12]  G. Kasieczka,et al.  How to GAN away Detector Effects , 2019, SciPost Physics.

[13]  J. Maalmi,et al.  Fast simulation of muons produced at the SHiP experiment using Generative Adversarial Networks , 2019 .

[14]  Patrick T. Komiske,et al.  OmniFold: A Method to Simultaneously Unfold All Observables. , 2019, Physical review letters.

[15]  M. Pierini,et al.  The DNNLikelihood: enhancing likelihood distribution with Deep Learning , 2019, The European Physical Journal C.

[16]  Eric A. Moreno,et al.  Interaction networks for the identification of boosted H→bb¯ decays , 2019, Physical Review D.

[17]  Eric A. Moreno,et al.  Interaction networks for the identification of boosted Higgs to bb decays , 2019 .

[18]  SHiP Collaboration.  Fast simulation of muons produced at the SHiP experiment using Generative Adversarial Networks , 2019, Journal of Instrumentation.

[19]  S. Carrazza,et al.  Lund jet images from generative and cycle-consistent adversarial networks , 2019, The European Physical Journal C.

[20]  Rashmish K. Mishra,et al.  Mass agnostic jet taggers , 2019, SciPost Physics.

[21]  G. Cottin,et al.  Boosted W and Z tagging with jet charge and deep learning , 2019, Physical Review D.

[22]  Atlas Collaboration.  Search for non-resonant Higgs boson pair production in the bbℓνℓν final state with the ATLAS detector in pp collisions at √s =13 TeV , 2019, 1908.06765.

[23]  Eric A. Moreno,et al.  JEDI-net: a jet identification algorithm based on interaction networks , 2019, The European Physical Journal C.

[24]  Stefan Wunsch,et al.  Reducing the Dependence of the Neural Network Function to Systematic Uncertainties in the Input Space , 2019, Computing and Software for Big Science.

[25]  B. Nachman,et al.  Neural networks for full phase-space reweighting and parameter tuning , 2019, Physical Review D.

[26]  S. Carrazza,et al.  Towards a new generation of parton densities with deep learning models , 2019, The European Physical Journal C.

[27]  Tilman Plehn,et al.  How to GAN LHC events , 2019, SciPost Physics.

[28]  Sofia Vallecorsa,et al.  3D convolutional GAN for fast simulation , 2019, EPJ Web of Conferences.

[29]  Sascha Diefenbacher,et al.  CapsNets continuing the convolutional quest , 2019, SciPost Physics.

[30]  C. Frye,et al.  binary junipr: An Interpretable Probabilistic Model for Discrimination. , 2019, Physical review letters.

[31]  M. Spannowsky,et al.  Adversarially-trained autoencoders for robust unsupervised new physics searches , 2019, Journal of High Energy Physics.

[32]  M. S. Albergo,et al.  Flow-based generative models for Markov chain Monte Carlo in lattice field theory , 2019, Physical Review D.

[33]  Jennifer Thompson,et al.  Deep-learning jets with uncertainties and more , 2019, SciPost Physics.

[34]  S. Mukherjee,et al.  Study of energy deposition patterns in hadron calorimeter for prompt and displaced jets using convolutional neural network , 2019, Journal of High Energy Physics.

[35]  Jernej F. Kamenik,et al.  Uncovering latent jet substructure , 2019, Physical Review D.

[36]  Sung Hak Lim,et al.  Interpretable deep learning for two-prong jet classification with jet spectra , 2019, Journal of High Energy Physics.

[37]  Nikita Kazeev,et al.  Cherenkov Detectors Fast Simulation Using Neural Networks , 2019, Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment.

[38]  S. Carrazza,et al.  Jet Grooming through Reinforcement Learning , 2019, Journal of Physics: Conference Series.

[39]  Sana Ketabchi Haghighat,et al.  DijetGAN: a Generative-Adversarial Network approach for the simulation of QCD dijet events at the LHC , 2019, Journal of High Energy Physics.

[40]  Benjamin Nachman,et al.  Machine learning templates for QCD factorization in the search for physics beyond the standard model , 2019, Journal of High Energy Physics.

[41]  T. Roy,et al.  A robust anomaly finder based on autoencoder , 2019, 1903.02032.

[42]  L. Gouskos,et al.  The Machine Learning landscape of top taggers , 2019, SciPost Physics.

[43]  Huilin Qu,et al.  ParticleNet: Jet Tagging via Particle Clouds , 2019, Physical Review D.

[44]  Yutaro Iiyama,et al.  Learning representations of irregular particle-detector geometry with distance-weighted graph networks , 2019, The European Physical Journal C.

[45]  B. Nachman,et al.  Automating the construction of jet observables with machine learning , 2019, Physical Review D.

[46]  Inkyu Park,et al.  Quark-Gluon Jet Discrimination Using Convolutional Neural Networks , 2019, Journal of the Korean Physical Society.

[47]  B. Nachman,et al.  Extending the search for new resonances with machine learning , 2019, Physical Review D.

[48]  Z. Li,et al.  Improving the measurement of the Higgs boson-gluon coupling using convolutional neural networks at e+e− colliders , 2019, Physical Review D.

[49]  Maurizio Pierini,et al.  LHC analysis-specific datasets with Generative Adversarial Networks , 2019, ArXiv.

[50]  M. Erdmann,et al.  Lorentz Boost Networks: autonomous physics-inspired feature engineering , 2018, Journal of Instrumentation.

[51]  Gregor Kasieczka,et al.  Quark-gluon tagging: Machine learning vs detector , 2018, SciPost Physics.

[52]  Dmitry Ulyanov,et al.  Generative Models for Fast Calorimeter Simulation: the LHCb case> , 2018, EPJ Web of Conferences.

[53]  Maria Spiropulu,et al.  Variational autoencoders for new physics mining at the Large Hadron Collider , 2018, Journal of High Energy Physics.

[54]  Jan M. Pawlowski,et al.  Reducing autocorrelation times in lattice simulations with generative adversarial networks , 2018, Mach. Learn. Sci. Technol..

[55]  David Lopez-Paz,et al.  Single-Model Uncertainties for Deep Learning , 2018, NeurIPS.

[56]  David Lopez-Paz,et al.  Frequentist uncertainty estimates for deep learning , 2018, ArXiv.

[57]  L. Pang,et al.  Regressive and generative neural networks for scalar field theory , 2018, Physical Review D.

[58]  L. Xia.  QBDT, a new boosting decision tree method with systematical uncertainties into training for High Energy Physics , 2018, Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment.

[59]  Maria Spiropulu,et al.  Pileup mitigation at the Large Hadron Collider with graph neural networks , 2018, The European Physical Journal Plus.

[60]  Patrick T. Komiske,et al.  Energy flow networks: deep sets for particle jets , 2018, Journal of High Energy Physics.

[61]  David Rousseau,et al.  Further developments of FORM , 2018, Journal of Physics: Conference Series.

[62]  Sven Krippendorf,et al.  GANs for generating EFT models , 2018, Physics Letters B.

[63]  S. Vallecorsa,et al.  Generative models for fast simulation , 2018, Journal of Physics: Conference Series.

[64]  P. Mendez Lorenzo,et al.  Three dimensional Generative Adversarial Networks for fast simulation , 2018, Journal of Physics: Conference Series.

[65]  S. V. Gleyzer,et al.  End-to-End Event Classification of High-Energy Physics Data , 2018, Journal of Physics: Conference Series.

[66]  D. Shih,et al.  Searching for new physics with deep autoencoders , 2018, Physical Review D.

[67]  Gregor Kasieczka,et al.  QCD or what? , 2018, SciPost Physics.

[68]  Enrico Bothmann,et al.  Reweighting a parton shower using a neural network: the final-state case , 2018, Journal of High Energy Physics.

[69]  V. S. Lugovsky,et al.  Review of Particle Physics , 2018, Physical Review D.

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

[71]  B. Nachman,et al.  Boosting H→bb¯$$ H\to b\overline{b} $$ with machine learning , 2018, Journal of High Energy Physics.

[72]  Tao Liu,et al.  Novelty Detection Meets Collider Physics , 2018, Physical Review D.

[73]  A. Simone,et al.  Guiding new physics searches with unsupervised learning , 2018, The European Physical Journal C.

[74]  Liam Moore,et al.  Reports of my demise are greatly exaggerated: $N$-subjettiness taggers take on jet images , 2018, SciPost Physics.

[75]  G. Salam,et al.  The Lund jet plane , 2018, Journal of High Energy Physics.

[76]  J. Monk,et al.  Deep learning as a parton shower , 2018, Journal of High Energy Physics.

[77]  Sung Hak Lim,et al.  Spectral analysis of jet substructure with neural networks: boosted Higgs case , 2018, Journal of High Energy Physics.

[78]  Martin Erdmann,et al.  Precise Simulation of Electromagnetic Calorimeter Showers Using a Wasserstein Generative Adversarial Network , 2018, Computing and Software for Big Science.

[79]  T. Trzciński,et al.  Generative Models for Fast Cluster Simulations in the TPC for the ALICE Experiment , 2018, Advances in Intelligent Systems and Computing.

[80]  Maria Spiropulu,et al.  Topology Classification with Deep Learning to Improve Real-Time Event Selection at the LHC , 2018, Computing and Software for Big Science.

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

[82]  Michela Paganini,et al.  Electromagnetic showers beyond shower shapes , 2018, 1806.05667.

[83]  Pablo de Castro,et al.  INFERNO: Inference-Aware Neural Optimisation , 2018, Comput. Phys. Commun..

[84]  R. D’Agnolo,et al.  Learning new physics from a machine , 2018, Physical Review D.

[85]  Suyong Choi,et al.  Infrared safety of a neural-net top tagging algorithm , 2018, Journal of High Energy Physics.

[86]  Deepak Kar,et al.  Unfolding with Generative Adversarial Networks , 2018, 1806.00433.

[87]  Jun Guo,et al.  Deep learning for R -parity violating supersymmetry searches at the LHC , 2018, Physical Review D.

[88]  B. Nachman,et al.  Anomaly Detection for Resonant New Physics with Machine Learning. , 2018, Physical review letters.

[89]  Francesco Pandolfi,et al.  Fast and Accurate Simulation of Particle Detectors Using Generative Adversarial Networks , 2018, Computing and Software for Big Science.

[90]  Gilles Louppe,et al.  Mining gold from implicit models to improve likelihood-free inference , 2018, Proceedings of the National Academy of Sciences.

[91]  Gilles Louppe,et al.  Constraining Effective Field Theories with Machine Learning. , 2018, Physical review letters.

[92]  Gilles Louppe,et al.  A guide to constraining effective field theories with machine learning , 2018, Physical Review D.

[93]  C. Frye,et al.  JUNIPR: a framework for unsupervised machine learning in particle physics , 2018, The European Physical Journal C.

[94]  M. Reece,et al.  Opening the black box of neural nets: case studies in stop/top discrimination , 2018, 1804.09278.

[95]  Song Han,et al.  Fast inference of deep neural networks in FPGAs for particle physics , 2018, Journal of Instrumentation.

[96]  Stefan Wunsch,et al.  Identifying the Relevant Dependencies of the Neural Network Response on Characteristics of the Input Space , 2018, Computing and Software for Big Science.

[97]  M. Schwartz,et al.  Jet charge and machine learning , 2018, Journal of High Energy Physics.

[98]  D. Shih,et al.  Pulling out all the tops with computer vision and deep learning , 2018, Journal of High Energy Physics.

[99]  Martin Erdmann,et al.  Generating and Refining Particle Detector Simulations Using the Wasserstein Distance in Adversarial Networks , 2018, Computing and Software for Big Science.

[100]  Patrick T. Komiske,et al.  Learning to classify from impure samples with high-dimensional data , 2018, Physical Review D.

[101]  Philip Harris,et al.  Machine learning uncertainties with adversarial neural networks , 2018, The European Physical Journal C.

[102]  Michela Paganini,et al.  CaloGAN: Simulating 3D High Energy Particle Showers in Multi-Layer Electromagnetic Calorimeters with Generative Adversarial Networks , 2017, ArXiv.

[103]  Kai Wang,et al.  Quark jet versus gluon jet: fully-connected neural networks with high-level features , 2017, Science China Physics, Mechanics & Astronomy.

[104]  Chang Wei Loh,et al.  Importance and construction of features in identifying new physics signals with deep learning , 2017, 1712.03806.

[105]  Alexander Glazov,et al.  Machine learning as an instrument for data unfolding , 2017, 1712.01814.

[106]  Shannon Egan,et al.  Long Short-Term Memory (LSTM) networks with jet constituents for boosted top tagging at the LHC , 2017, ArXiv.

[107]  Prabhat,et al.  Deep Neural Networks for Physics Analysis on low-level whole-detector data at the LHC , 2017, Journal of Physics: Conference Series.

[108]  B. Nachman,et al.  Convolved substructure: analytically decorrelating jet substructure observables , 2017, 1710.06859.

[109]  A. Larkoski,et al.  Novel jet observables from machine learning , 2017, 1710.01305.

[110]  Fedor Ratnikov,et al.  Towards automation of data quality system for CERN CMS experiment , 2017, ArXiv.

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

[112]  J. A. Aguilar-Saavedra,et al.  A generic anti-QCD jet tagger , 2017, 1709.01087.

[113]  B. Nachman,et al.  Classification without labels: learning from mixed samples in high energy physics , 2017, Journal of High Energy Physics.

[114]  I. A. Monroy,et al.  Observation of the Decays Λ_{b}^{0}→χ_{c1}pK^{-} and Λ_{b}^{0}→χ_{c2}pK^{-}. , 2017, Physical review letters.

[115]  Gregor Kasieczka,et al.  Deep-learned Top Tagging with a Lorentz Layer , 2017, SciPost Physics.

[116]  Patrick T. Komiske,et al.  Pileup Mitigation with Machine Learning (PUMML) , 2017, Journal of High Energy Physics.

[117]  M. Freytsis,et al.  (Machine) learning to do more with less , 2017, Journal of High Energy Physics.

[118]  Zihao Jiang,et al.  Identification of Jets Containing b-Hadrons with Recurrent Neural Networks at the ATLAS Experiment , 2017 .

[119]  Benjamin Nachman,et al.  Accelerating Science with Generative Adversarial Networks: An Application to 3D Particle Showers in Multilayer Calorimeters. , 2017, Physical review letters.

[120]  A. Larkoski,et al.  How much information is in a jet? , 2017, Journal of High Energy Physics.

[121]  I. Babuschkin,et al.  Observation of the decays $\Lambda_b^0 \to \chi_{c1} p K^-$ and $\Lambda_b^0 \to \chi_{c2} p K^-$ , 2017, 1704.07900.

[122]  Wojciech Fedorko,et al.  Jet Constituents for Deep Neural Network Based Top Quark Tagging , 2017, ArXiv.

[123]  Pierre Baldi,et al.  Decorrelated jet substructure tagging using adversarial neural networks , 2017, Physical Review D.

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

[125]  G. Kasieczka,et al.  Deep-learning top taggers or the end of QCD? , 2017, 1701.08784.

[126]  Luke de Oliveira,et al.  Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis , 2017, Computing and Software for Big Science.

[127]  Patrick T. Komiske,et al.  Deep learning in color: towards automated quark/gluon jet discrimination , 2016, Journal of High Energy Physics.

[128]  Gilles Louppe,et al.  Learning to Pivot with Adversarial Networks , 2016, NIPS.

[129]  B. Nachman.  Investigating the Quantum Properties of Jets and the Search for a Supersymmetric Top Quark Partner with the ATLAS Detector , 2016, 1609.03242.

[130]  E. Dawe,et al.  Parton Shower Uncertainties in Jet Substructure Analyses with Deep Neural Networks , 2016, 1609.00607.

[131]  A. Rogozhnikov,et al.  Reweighting with Boosted Decision Trees , 2016, 1608.05806.

[132]  Khachatryan,et al.  Jet energy scale and resolution in the CMS experiment in pp collisions at 8 TeV , 2016, 1607.03663.

[133]  P. Baldi,et al.  Jet Substructure Classification in High-Energy Physics with Deep Neural Networks , 2016, 1603.09349.

[134]  P. Harris,et al.  Thinking outside the ROCs: Designing Decorrelated Taggers (DDT) for jet substructure , 2016, 1603.00027.

[135]  Pierre Baldi,et al.  Parameterized neural networks for high-energy physics , 2016, The European Physical Journal C.

[136]  Sepp Hochreiter,et al.  Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs) , 2015, ICLR.

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

[138]  Gilles Louppe,et al.  Approximating Likelihood Ratios with Calibrated Discriminative Classifiers , 2015, 1506.02169.

[139]  J. T. Childers,et al.  Jet energy measurement and its systematic uncertainty in proton–proton collisions at $$\sqrt{s}=7$$s=7 TeV with the ATLAS detector , 2015 .

[140]  J. T. Childers,et al.  Jet energy measurement and its systematic uncertainty in proton–proton collisions at \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \be , 2015, The European Physical Journal C.

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

[142]  P. Baldi,et al.  Searching for exotic particles in high-energy physics with deep learning , 2014, Nature Communications.

[143]  Mike Williams,et al.  uBoost: a boosting method for producing uniform selection efficiencies from multivariate classifiers , 2013, 1305.7248.

[144]  C. Lester,et al.  Significance Variables , 2013, 1303.7009.

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

[146]  D Martschei,et al.  Advanced event reweighting using multivariate analysis , 2012 .

[147]  K. Cranmer,et al.  Power-Constrained Limits , 2011, 1105.3166.

[148]  K. Cranmer,et al.  Asymptotic formulae for likelihood-based tests of new physics , 2010, 1007.1727.

[149]  Nikolai Gagunashvili,et al.  Machine learning approach to inverse problem and unfolding procedure , 2010, 1004.2006.

[150]  A. Kiureghian,et al.  Aleatory or epistemic? Does it matter? , 2009 .

[151]  Y. Meng,et al.  Applying Bayesian neural networks to event reconstruction in reactor neutrino experiments , 2007, 0712.4042.

[152]  M. Wolter,et al.  TMVA - Toolkit for Multivariate Data Analysis , 2007, physics/0703039.

[153]  A. Read.  Presentation of search results: the CLs technique , 2002 .

[154]  T. Junk,et al.  Confidence Level Computation for Combining Searches with Small Statistics , 1999, hep-ex/9902006.

[155]  Charles M. Bishop.  Bayesian Neural Networks , 1997, J. Braz. Comput. Soc..

[156]  Alan D. Martin,et al.  Review of Particle Physics , 2000, Physical Review D.

[157]  Thorsteinn S. Rögnvaldsson,et al.  Using neural networks to identify jets , 1991 .

[158]  R. Barlow,et al.  Extended maximum likelihood , 1990 .

[159]  Igor Kononenko,et al.  Bayesian neural networks , 1989, Biological Cybernetics.

[160]  M. L. Stevenson,et al.  PION-PION INTERACTION IN THE REACTION /anti p/ + p $Yields$ 2$pi$$sup +$ + 2$pi$$sup -$ + n$pi$$sup 0$ , 1962 .

[161]  S. S. Wilks.  The Large-Sample Distribution of the Likelihood Ratio for Testing Composite Hypotheses , 1938 .

[162]  B. Nachman,et al.  JHEP Boosting H → bb̄ with Machine Learning , 2018 .

[163]  Heavy flavor identification at CMS with deep neural networks , 2017 .

[164]  F. Taylor,et al.  Jet energy measurement and its systematic uncertainty in proton-proton collisions at √s=7 TeV with the ATLAS detector , 2015 .

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

[166]  J. T. Childers,et al.  Observation of a new particle in the search for the Standard Model Higgs boson with the ATLAS detector at the LHC , 2012 .

[167]  S. Saucedo.  Bayesian Neural Networks for Classification , 2006 .

[168]  David Mackay,et al.  Probable networks and plausible predictions - a review of practical Bayesian methods for supervised neural networks , 1995 .

[169]  B. Efron.  Bootstrap Methods: Another Look at the Jackknife , 1979 .

[170]  M. L. Stevenson,et al.  Pion-Pion Interaction in the Reaction p ¯ + p → 2 π + + 2 π − + n π 0 , 1962 .

[171]  E. S. Pearson,et al.  On the Problem of the Most Efficient Tests of Statistical Hypotheses , 1933 .

[172]  Andy Davis,et al.  This Paper Is Included in the Proceedings of the 12th Usenix Symposium on Operating Systems Design and Implementation (osdi '16). Tensorflow: a System for Large-scale Machine Learning Tensorflow: a System for Large-scale Machine Learning , 2022 .