Machine Learning under Resource Constraints, Volume 2: Discovery in Physics

364 Pages·2022·44.029 MB·English

by  Katharina Morik, Wolfgang Rhode, Jens Buß, Andreas Becker

Katharina Morik, Wolfgang Rhode (Eds.)
Machine Learning under Resource Constraints · Discovery in Physics
Also of interest
Volume 1
Machine Learning under Resource Constraints.
Fundamentals
Morik, Marwedel (Eds.), 2023
ISBN 978-3-11-078593-7, e-ISBN 978-3-11-078594-4
Volume 3
Machine Learning under Resource Constraints.
Applications
Morik, Rahnenführer, Wietfeld (Eds.), 2023
ISBN 978-3-11-078597-5, e-ISBN 978-3-11-078598-2
Machine Learning
under Resource
Constraints
Final Report of CRC 876
Editor in Chief
Katharina Morik
Volume 2/3

Machine Learning
under Resource
Constraints
Discovery in Physics
Edited by
Katharina Morik and Wolfgang Rhode
Editors
Prof. Dr. Katharina Morik
TU Dortmund University
Department of Computer Sciences
Chair for Artificial Intelligence
Computer Science 8
Otto-Hahn-Str. 12
44221 Dortmund
Germany
Prof. Dr. Dr. Wolfgang Rhode
TU Dortmund University
Department of Physics
Chair for Experimental Physics 5b
Otto-Hahn-Str. 4b
44221 Dortmund
Germany
ISBN 978-3-11-078595-1
e-ISBN (PDF) 978-3-11-078596-8
e-ISBN (EPUB) 978-3-11-078613-2
DOI https://doi.org/10.1515/9783110785968
This work is licensed under the Creative Commons Attribution 4.0 International License. For details
go to https://creativecommons.org/licenses/by/4.0/.
Creative Commons license terms for re-use do not apply to any content (such as graphs, figures,
photos, excerpts, etc.) not original to the Open Access publication and further permission may be
required from the rights holder. The obligation to research and clear permission lies solely with the
party re-using the material.
Library of Congress Control Number: 2022949268
Bibliographic information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie;
detailed bibliographic data are available on the Internet at http://dnb.dnb.de.
© 2023 with the author(s), editing © 2023 Katharina Morik and Wolfgang Rhode,
published by Walter de Gruyter GmbH, Berlin/Boston
This book is published open access at www.degruyter.com.
Cover image: Collaborative Research Center 876
Printing and binding: CPI books GmbH, Leck
www.degruyter.com
|
Dedication:WededicatethisbooktothememoryofourfriendandcolleagueBernhard
Spaan,whowalkedthispathbetweenphysicsandcomputersciencewithusformany
yearsuntilhesadlyleftusduringtheworkonthisbook.

Contents
1 Introduction 1
|
1.1 Basics,Questions,andMotivation|1
1.2 MachineLearningasModeloftheScientificProcess|4
1.2.1 EarlyApproachestoScientificDiscovery|4
1.2.2 KnowledgeRepresentation–WhatisaTheory?|5
1.2.3 TowardsProbabilities|7
1.2.4 TheBigDataMove|8
1.3 FromthePhysicalTheorytothePhysicalObservable|10
1.4 FromtheMeasuredVariabletotheMeasuringPoint|18
1.5 TheEpistemologyofPhysicsinConcertwithMachineLearning|26
1.5.1 FurtherReadinginthisBook|28
2 ChallengesinParticleandAstroparticlePhysics 31
|
2.1 PhysicalMotivation,Problems,andExamples|31
2.2 AstroparticlePhysics|32
2.2.1 Experiments|34
2.3 ParticlePhysics|42
2.3.1 Experiments|42
3 KeyConceptsinMachineLearningandDataAnalysis 51
|
3.1 OverviewoftheFieldofMachineLearning|51
3.1.1 LearningTasks|52
3.1.2 ProcessingParadigmsofMachineLearning|55
3.1.3 MachineLearningPipelines|56
3.1.4 MinimumRedundancyMaximumRelevance(MRMR)|58
3.2 Optimization|59
3.2.1 StochasticGradientDescent|60
3.2.2 Newton-RaphsonOptimization|61
3.3 TheoriesofMachineLearning|62
3.3.1 ComputationalLearningTheory|64
3.4 TreeModels|66
3.4.1 EnsembleMethods|67
3.4.2 ImplementationsandHardwareConsiderations|69
3.5 NeuralNetworks|70
3.5.1 ArchitecturesofDNNs|71
3.5.2 RobustnessofDNNs|73
X | Contents
3.5.3 DeepLearningTheory|73
3.5.4 Explanations|74
3.5.5 HardwareConsiderations|75
4 DataAcquisitionandDataStructure 77
|
4.1 Introduction|77
4.2 DataAcquisition|79
4.2.1 DataAcquisitionforLHCb|79
4.2.2 DataAcquisitionforImagingAirCherenkovTelescopes|82
4.2.3 DataAcquisitionforIceCube|85
4.3 DataStructures|89
4.3.1 DataStructuresforLHCb|89
4.3.2 DataStructuresforIACTs|92
4.3.3 DataStructuresforIceCube|96
4.4 GPU-BasedTriggerDecisions|98
5 MonteCarloSimulations 103
|
5.1 LHCb:MonteCarloSimulationsandLibrariesinParticlePhysics|103
5.1.1 Astro:MonteCarloSimulations,Libraries|105
5.2 SimulationEfficiencyStudies|108
5.2.1 Corsika–ActiveLearning|108
5.2.2 ControloftheSimulation–ActiveSampling|112
5.2.3 Corsika8NewModularLibrary|118
5.2.4 ARM-ClusterforCorsika|120
5.3 ValidationoftheSimulation|126
5.3.1 Introduction|126
5.3.2 MismatchesBetweenObservedandSimulatedData|127
5.3.3 DetectionofMismatches|128
5.4 Keynote:TheMuonPuzzle|133
5.4.1 Introduction|133
5.4.2 Meta-AnalysisofMuonMeasurementsinAirShowers|134
5.4.3 MuonProductioninAirShowers|136
5.4.4 RelatedMeasurementsattheLHCatCERN|138
5.4.5 Fixed-TargetExperimentsatSPSandLHC|141
5.4.6 SummaryandOutlook|142
6 DataStorageandAccess 145
|
6.1 Introduction|145
6.2 ResearchDataManagement|145