Search for New Phenomena in Two-Body Invariant Mass Distributions Using Unsupervised Machine Learning for Anomaly Detection at √s = 13 TeV with the ATLAS Detector

G. Aad; T.P.A. Åkesson; E.E. Corrigan; C. Doglioni; J. Geisen; E. Hansen; V. Hedberg; Hannah Herde; B. Konya; E. Lytken; R. Poettgen; N.D. Simpson; O. Smirnova; L. Zwalinski; ATLAS Collaboration

doi:10.1103/PhysRevLett.132.081801

Search for New Phenomena in Two-Body Invariant Mass Distributions Using Unsupervised Machine Learning for Anomaly Detection at √s = 13 TeV with the ATLAS Detector

G. Aad, T.P.A. Åkesson, E.E. Corrigan, C. Doglioni, J. Geisen, E. Hansen, V. Hedberg, Hannah Herde, B. Konya, E. Lytken, R. Poettgen, N.D. Simpson, O. Smirnova, L. Zwalinski, ATLAS Collaboration

Research output: Contribution to journal › Article › peer-review

Abstract

Searches for new resonances are performed using an unsupervised anomaly-detection technique. Events with at least one electron or muon are selected from 140 fb−1 of pp collisions at √s ¼ 13 TeV recorded by ATLAS at the Large Hadron Collider. The approach involves training an autoencoder on data, and subsequently defining anomalous regions based on the reconstruction loss of the decoder. Studies focus on nine invariant mass spectra that contain pairs of objects consisting of one light jet or b jet and either one lepton (e; μ), photon, or second light jet or b jet in the anomalous regions. No significant deviations from the background hypotheses are observed. Limits on contributions from generic Gaussian signals with various widths of the resonance mass are obtained for nine invariant masses in the anomalous regions. © 2024 CERN, for the ATLAS Collaboration.

Original language	English
Article number	081801
Journal	Physical Review Letters
Volume	132
Issue number	8
DOIs	https://doi.org/10.1103/PhysRevLett.132.081801
Publication status	Published - 2024

Bibliographical note

Number of authors = 2909

EID = 85186742137

Article no = 081801

Affiliation = Aad G., CPPM, Aix-Marseille Université, CNRS, IN2P3, Marseille, France

Affiliation = Zou W., Nevis Laboratory, Columbia University, Irvington, NY, United States

Affiliation = Zwalinski L., CERN, Geneva, Switzerland

Subject classification (UKÄ)

Subatomic Physics

Free keywords

Anomaly detection
Machine learning
Tellurium compounds
Anomalous regions
ATLAS detectors
Auto encoders
Invariant mass distribution
Large Hadron Collider
Large-hadron colliders
Region-based
Unsupervised anomaly detection
Unsupervised machine learning
Mass spectrometry

Access to Document

10.1103/PhysRevLett.132.081801Licence: CC BY

Cite this

Aad, G., Åkesson, T. P. A., Corrigan, E. E., Doglioni, C., Geisen, J., Hansen, E., Hedberg, V., Herde, H., Konya, B., Lytken, E., Poettgen, R., Simpson, N. D., Smirnova, O., Zwalinski, L., & ATLAS Collaboration (2024). Search for New Phenomena in Two-Body Invariant Mass Distributions Using Unsupervised Machine Learning for Anomaly Detection at √s = 13 TeV with the ATLAS Detector. Physical Review Letters, 132(8), Article 081801. https://doi.org/10.1103/PhysRevLett.132.081801

@article{f0f650f1831f4456adacb898d2580c3b,

title = "Search for New Phenomena in Two-Body Invariant Mass Distributions Using Unsupervised Machine Learning for Anomaly Detection at √s = 13 TeV with the ATLAS Detector",

abstract = "Searches for new resonances are performed using an unsupervised anomaly-detection technique. Events with at least one electron or muon are selected from 140 fb−1 of pp collisions at √s ¼ 13 TeV recorded by ATLAS at the Large Hadron Collider. The approach involves training an autoencoder on data, and subsequently defining anomalous regions based on the reconstruction loss of the decoder. Studies focus on nine invariant mass spectra that contain pairs of objects consisting of one light jet or b jet and either one lepton (e; μ), photon, or second light jet or b jet in the anomalous regions. No significant deviations from the background hypotheses are observed. Limits on contributions from generic Gaussian signals with various widths of the resonance mass are obtained for nine invariant masses in the anomalous regions. {\textcopyright} 2024 CERN, for the ATLAS Collaboration.",

keywords = "Anomaly detection, Machine learning, Tellurium compounds, Anomalous regions, ATLAS detectors, Auto encoders, Invariant mass distribution, Large Hadron Collider, Large-hadron colliders, Region-based, Unsupervised anomaly detection, Unsupervised machine learning, Mass spectrometry",

author = "G. Aad and T.P.A. {\AA}kesson and E.E. Corrigan and C. Doglioni and J. Geisen and E. Hansen and V. Hedberg and Hannah Herde and B. Konya and E. Lytken and R. Poettgen and N.D. Simpson and O. Smirnova and L. Zwalinski and {ATLAS Collaboration}",

note = "Number of authors = 2909 EID = 85186742137 Article no = 081801 Affiliation = Aad G., CPPM, Aix-Marseille Universit{\'e}, CNRS, IN2P3, Marseille, France Affiliation = Zou W., Nevis Laboratory, Columbia University, Irvington, NY, United States Affiliation = Zwalinski L., CERN, Geneva, Switzerland",

year = "2024",

doi = "10.1103/PhysRevLett.132.081801",

language = "English",

volume = "132",

journal = "Physical Review Letters",

issn = "0031-9007",

publisher = "American Physical Society",

number = "8",

}

Aad, G, Åkesson, TPA, Corrigan, EE, Doglioni, C, Geisen, J, Hansen, E, Hedberg, V , Herde, H , Konya, B , Lytken, E , Poettgen, R, Simpson, ND, Smirnova, O, Zwalinski, L & ATLAS Collaboration 2024, 'Search for New Phenomena in Two-Body Invariant Mass Distributions Using Unsupervised Machine Learning for Anomaly Detection at √s = 13 TeV with the ATLAS Detector', Physical Review Letters, vol. 132, no. 8, 081801. https://doi.org/10.1103/PhysRevLett.132.081801

TY - JOUR

T1 - Search for New Phenomena in Two-Body Invariant Mass Distributions Using Unsupervised Machine Learning for Anomaly Detection at √s = 13 TeV with the ATLAS Detector

AU - Aad, G.

AU - Åkesson, T.P.A.

AU - Corrigan, E.E.

AU - Doglioni, C.

AU - Geisen, J.

AU - Hansen, E.

AU - Hedberg, V.

AU - Herde, Hannah

AU - Konya, B.

AU - Lytken, E.

AU - Poettgen, R.

AU - Simpson, N.D.

AU - Smirnova, O.

AU - Zwalinski, L.

AU - ATLAS Collaboration

N1 - Number of authors = 2909 EID = 85186742137 Article no = 081801 Affiliation = Aad G., CPPM, Aix-Marseille Université, CNRS, IN2P3, Marseille, France Affiliation = Zou W., Nevis Laboratory, Columbia University, Irvington, NY, United States Affiliation = Zwalinski L., CERN, Geneva, Switzerland

PY - 2024

Y1 - 2024

N2 - Searches for new resonances are performed using an unsupervised anomaly-detection technique. Events with at least one electron or muon are selected from 140 fb−1 of pp collisions at √s ¼ 13 TeV recorded by ATLAS at the Large Hadron Collider. The approach involves training an autoencoder on data, and subsequently defining anomalous regions based on the reconstruction loss of the decoder. Studies focus on nine invariant mass spectra that contain pairs of objects consisting of one light jet or b jet and either one lepton (e; μ), photon, or second light jet or b jet in the anomalous regions. No significant deviations from the background hypotheses are observed. Limits on contributions from generic Gaussian signals with various widths of the resonance mass are obtained for nine invariant masses in the anomalous regions. © 2024 CERN, for the ATLAS Collaboration.

AB - Searches for new resonances are performed using an unsupervised anomaly-detection technique. Events with at least one electron or muon are selected from 140 fb−1 of pp collisions at √s ¼ 13 TeV recorded by ATLAS at the Large Hadron Collider. The approach involves training an autoencoder on data, and subsequently defining anomalous regions based on the reconstruction loss of the decoder. Studies focus on nine invariant mass spectra that contain pairs of objects consisting of one light jet or b jet and either one lepton (e; μ), photon, or second light jet or b jet in the anomalous regions. No significant deviations from the background hypotheses are observed. Limits on contributions from generic Gaussian signals with various widths of the resonance mass are obtained for nine invariant masses in the anomalous regions. © 2024 CERN, for the ATLAS Collaboration.

KW - Anomaly detection

KW - Machine learning

KW - Tellurium compounds

KW - Anomalous regions

KW - ATLAS detectors

KW - Auto encoders

KW - Invariant mass distribution

KW - Large Hadron Collider

KW - Large-hadron colliders

KW - Region-based

KW - Unsupervised anomaly detection

KW - Unsupervised machine learning

KW - Mass spectrometry

U2 - 10.1103/PhysRevLett.132.081801

DO - 10.1103/PhysRevLett.132.081801

M3 - Article

C2 - 38457710

SN - 0031-9007

VL - 132

JO - Physical Review Letters

JF - Physical Review Letters

IS - 8

M1 - 081801

ER -

Search for New Phenomena in Two-Body Invariant Mass Distributions Using Unsupervised Machine Learning for Anomaly Detection at √s = 13 TeV with the ATLAS Detector

Abstract

Bibliographical note

Subject classification (UKÄ)

Free keywords

Access to Document

Fingerprint

Cite this