Mass agnostic jet taggers

Bradshaw, Layne; Mishra, Rashmish K.; Mitridate, Andrea; Ostdiek, Bryan

doi:10.21468/SciPostPhys.8.1.011

Title: Mass agnostic jet taggers

Abstract

Searching for new physics in large data sets needs a balance between two competing effects—signal identification vs background distortion. In this work, we perform a systematic study of both single variable and multivariate jet tagging methods that aim for this balance. The methods preserve the shape of the background distribution by either augmenting the training procedure or the data itself. Multiple quantitative metrics to compare the methods are considered, for tagging 2-, 3-, or 4-prong jets from the QCD background. This is the first study to show that the data augmentation techniques of Planing and PCA based scaling deliver similar performance as the augmented training techniques of Adversarial NN and uBoost, but are both easier to implement and computationally cheaper.

Authors:

Bradshaw, Layne ^[1]; Mishra, Rashmish K. ^[2]; Mitridate, Andrea ^[2]; Ostdiek, Bryan ^[1]

University of Oregon
National Institute for Nuclear Physics, Scuola Normale Superiore di Pisa

Publication Date:: Fri Jan 24 00:00:00 EST 2020

Research Org.:: Univ. of Oregon, Eugene, OR (United States)

Sponsoring Org.:: USDOE Office of Science (SC)

OSTI Identifier:: 1594194

Alternate Identifier(s):: OSTI ID: 1802234; OSTI ID: 1905954

Grant/Contract Number:: SC0011640

Resource Type:: Published Article

Journal Name:: SciPost Physics Proceedings

Additional Journal Information:: Journal Name: SciPost Physics Proceedings Journal Volume: 8 Journal Issue: 1; Journal ID: ISSN 2542-4653

Publisher:: Stichting SciPost

Country of Publication:: Netherlands

Language:: English

Subject:: 71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSICS; Physics

Citation Formats


                    Bradshaw, Layne, Mishra, Rashmish K., Mitridate, Andrea, and Ostdiek, Bryan. Mass agnostic jet taggers.  Netherlands: N. p., 2020. 
Web.  doi:10.21468/SciPostPhys.8.1.011.

Copy to clipboard


                    Bradshaw, Layne, Mishra, Rashmish K., Mitridate, Andrea, & Ostdiek, Bryan. Mass agnostic jet taggers.  Netherlands.  https://doi.org/10.21468/SciPostPhys.8.1.011

Copy to clipboard


                    Bradshaw, Layne, Mishra, Rashmish K., Mitridate, Andrea, and Ostdiek, Bryan. Fri .  
"Mass agnostic jet taggers".  Netherlands.  https://doi.org/10.21468/SciPostPhys.8.1.011.

Copy to clipboard


                    
@article{osti_1594194,

  title        = {Mass agnostic jet taggers},

  author       = {Bradshaw, Layne and Mishra, Rashmish K. and Mitridate, Andrea and Ostdiek, Bryan},

  abstractNote = {Searching for new physics in large data sets needs a balance between two competing effects—signal identification vs background distortion. In this work, we perform a systematic study of both single variable and multivariate jet tagging methods that aim for this balance. The methods preserve the shape of the background distribution by either augmenting the training procedure or the data itself. Multiple quantitative metrics to compare the methods are considered, for tagging 2-, 3-, or 4-prong jets from the QCD background. This is the first study to show that the data augmentation techniques of Planing and PCA based scaling deliver similar performance as the augmented training techniques of Adversarial NN and uBoost, but are both easier to implement and computationally cheaper.},

  doi          = {10.21468/SciPostPhys.8.1.011},

  journal      = {SciPost Physics Proceedings},

  number       = 1,

  volume       = 8,

  place        = {Netherlands},

  year         = {Fri Jan 24 00:00:00 EST 2020},

  month        = {Fri Jan 24 00:00:00 EST 2020}

}

Copy to clipboard

Journal Article:

Free Publicly Available Full Text

Publisher's Version of Record
https://doi.org/10.21468/SciPostPhys.8.1.011

Other availability

Search WorldCat to find libraries that may hold this journal

Citation Metrics:

Cited by: 28 works

Citation information provided by
Web of Science

Save / Share:

Export Metadata

Save to My Library

Works referenced in this record:

uBoost: a boosting method for producing uniform selection efficiencies from multivariate classifiers
journal, December 2013

Stevens, J.; Williams, M.
Journal of Instrumentation, Vol. 8, Issue 12
DOI: 10.1088/1748-0221/8/12/P12013

Parameterized neural networks for high-energy physics
journal, April 2016

Baldi, Pierre; Cranmer, Kyle; Faucett, Taylor
The European Physical Journal C, Vol. 76, Issue 5
DOI: 10.1140/epjc/s10052-016-4099-4

A brief introduction to PYTHIA 8.1
journal, June 2008

Sjöstrand, Torbjörn; Mrenna, Stephen; Skands, Peter
Computer Physics Communications, Vol. 178, Issue 11
DOI: 10.1016/j.cpc.2008.01.036

Anomaly Detection for Resonant New Physics with Machine Learning
journal, December 2018

Collins, Jack; Howe, Kiel; Nachman, Benjamin
Physical Review Letters, Vol. 121, Issue 24
DOI: 10.1103/PhysRevLett.121.241803

Dedicated strategies for triboson signals from cascade decays of vector resonances
journal, April 2019

Agashe, Kaustubh; Collins, Jack H.; Du, Peizhi
Physical Review D, Vol. 99, Issue 7
DOI: 10.1103/PhysRevD.99.075016

New approaches for boosting to uniformity
journal, March 2015

Rogozhnikov, A.; Bukva, A.; Gligorov, V.
Journal of Instrumentation, Vol. 10, Issue 03
DOI: 10.1088/1748-0221/10/03/T03002

Jet substructure at the Large Hadron Collider: A review of recent advances in theory and machine learning
journal, November 2019

Larkoski, Andrew J.; Moult, Ian; Nachman, Benjamin
Physics Reports
DOI: 10.1016/j.physrep.2019.11.001

Deep Learning and Its Application to LHC Physics
journal, October 2018

Guest, Dan; Cranmer, Kyle; Whiteson, Daniel
Annual Review of Nuclear and Particle Science, Vol. 68, Issue 1
DOI: 10.1146/annurev-nucl-101917-021019

Stealth multiboson signals
journal, October 2017

Aguilar-Saavedra, J. A.
The European Physical Journal C, Vol. 77, Issue 10
DOI: 10.1140/epjc/s10052-017-5289-4

Similar Records in DOE PAGES and OSTI.GOV collections:

B-jet and c-jet identification with Neural Networks as well as combination of multivariate analyses for the search for of multivariate analyses for the search for single top-quark production

Technical Report Renz, Manuel ; /Karlsruhe U., EKP

In the first part of this diploma thesis, the current version of the KIT Flavor Separator, a neural network which is able to distinguish between tagged b-quark jets and tagged c/light-quark jets, is presented. In comparison with previous versions four new input variables are utilized and new Monte Carlo samples with a larger number of simulated events are used for the training of the neural network. It is illustrated that the output of the neural network is continuously distributed between 1 and -1, whereas b-quark jets accumulate at 1, however, c-quark jets and light-quark jets have outputs next to -1.more »« less
https://doi.org/10.2172/957074

Full Text Available
Do graph neural networks learn traditional jet substructure?

Conference Mokhtar, Farouk ; Kansal, Raghav ; Duarte, Javier

At the CERN LHC, the task of jet tagging, whose goal is to infer the origin of a jet given a set of final-state particles, is dominated by machine learning methods. Graph neural networks have been used to address this task by treating jets as point clouds with underlying, learnable, edge connections between the particles inside. We explore the decision-making process for one such state-of-the-art network, ParticleNet, by looking for relevant edge connections identified using the layerwise-relevance propagation technique. As the model is trained, we observe changes in the distribution of relevant edges connecting different intermediate clusters of particles, knownmore »« less
Full Text Available
A neural jet charge tagger for the measurement of the B$$0\atop{s}$$-$$\bar{B}$$$$0\atop{s}$$ oscillation frequency at CDF

Thesis/Dissertation Lecci, Claudia

A Jet Charge Tagger algorithm for b-flavour tagging for the measurement of Δm_s at CDF has been presented. The tagger is based on a b-track probability variable and a b-jet probability variable, both obtained by combining the information available in bmore »« less
https://doi.org/10.2172/911837

Full Text Available
Boosting $$ H\to b\overline{b} $$ with machine learning

Journal Article Lin, Joshua ; Freytsis, Marat ; Moult, Ian ; ... - Journal of High Energy Physics (Online)

High-p_T Higgs production at hadron colliders provides a direct probe of the internal structure of the gg → H loop with the H→more »« less
Cited by 47
https://doi.org/10.1007/JHEP10(2018)101

Full Text Available
Introduction to HOBIT, a b-jet identification tagger at the CDF experiment optimized for light Higgs boson searches

Journal Article Freeman, J. ; Junk, T. ; Kirby, M. ; ... - Nuclear Instruments and Methods in Physics Research. Section A, Accelerators, Spectrometers, Detectors and Associated Equipment

We present the development and validation of the Higgs Optimized b Identification Tagger (HOBIT), a multivariate b-jet identification algorithm optimized for Higgs boson searches at the CDF experiment at the Fermilab Tevatron. At collider experiments, b taggers allow one to distinguish particle jets containing B hadrons from other jets; these algorithms have been used for many years with great success at CDF. HOBIT has been designed specifically for use in searches for light Higgs bosons decaying via H ! b\bar{b}. This fact combined with the extent to which HOBIT synthesizes and extends the best ideas of previous taggers makes HOBITmore »« less
https://doi.org/10.1016/j.nima.2012.09.021

Full Text Available

Similar Records

Title: Mass agnostic jet taggers

Abstract

Citation Formats

uBoost: a boosting method for producing uniform selection efficiencies from multivariate classifiers journal, December 2013

Parameterized neural networks for high-energy physics journal, April 2016

A brief introduction to PYTHIA 8.1 journal, June 2008

Anomaly Detection for Resonant New Physics with Machine Learning journal, December 2018

Dedicated strategies for triboson signals from cascade decays of vector resonances journal, April 2019

New approaches for boosting to uniformity journal, March 2015

Jet substructure at the Large Hadron Collider: A review of recent advances in theory and machine learning journal, November 2019

Deep Learning and Its Application to LHC Physics journal, October 2018

Stealth multiboson signals journal, October 2017

uBoost: a boosting method for producing uniform selection efficiencies from multivariate classifiers
journal, December 2013

Parameterized neural networks for high-energy physics
journal, April 2016

A brief introduction to PYTHIA 8.1
journal, June 2008

Anomaly Detection for Resonant New Physics with Machine Learning
journal, December 2018

Dedicated strategies for triboson signals from cascade decays of vector resonances
journal, April 2019

New approaches for boosting to uniformity
journal, March 2015

Jet substructure at the Large Hadron Collider: A review of recent advances in theory and machine learning
journal, November 2019

Deep Learning and Its Application to LHC Physics
journal, October 2018

Stealth multiboson signals
journal, October 2017