Search for new phenomena in two-body invariant mass distributions using unsupervised machine learning for anomaly detection at $\sqrt{s} = 13$ TeV with the ATLAS detector

The ATLAS collaboration
Phys.Rev.Lett. 132 (2024) 081801, 2024.

Abstract (data 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 $\textit{p} \textit{p}$ collisions at $\sqrt{s}$ = 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 $\textit{b}$-jet and either one lepton ($\textit{e}, \mu$), photon, or second light jet or $\textit{b}$-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.

  • fig_01

    Data from Figure 01

    10.17182/hepdata.144864.v1/t1

    Distributions of the anomaly score from the AE for data and five benchmark BSM models. Their legends, from top to...

  • fig_02

    Data from Figure 02

    10.17182/hepdata.144864.v1/t2

    Invariant mass distributions of jet+Y for $M_{jY}$ > 0.3 TeV in the 10 pb AR along with the fit of...

  • fig_03_jj

    Data from Figure 03

    10.17182/hepdata.144864.v1/t3

    Values of $\Delta Z$ for the discovery sensitivity, as defined in the text, as a function of the invariant mass...

  • fig_03_jb

    Data from Figure 03

    10.17182/hepdata.144864.v1/t4

    Values of $\Delta Z$ for the discovery sensitivity, as defined in the text, as a function of the invariant mass...

  • fig_03_bb

    Data from Figure 03

    10.17182/hepdata.144864.v1/t5

    Values of $\Delta Z$ for the discovery sensitivity, as defined in the text, as a function of the invariant mass...

  • fig_03_je

    Data from Figure 03

    10.17182/hepdata.144864.v1/t6

    Values of $\Delta Z$ for the discovery sensitivity, as defined in the text, as a function of the invariant mass...

  • fig_03_be

    Data from Figure 03

    10.17182/hepdata.144864.v1/t7

    Values of $\Delta Z$ for the discovery sensitivity, as defined in the text, as a function of the invariant mass...

  • fig_03_jg

    Data from Figure 03

    10.17182/hepdata.144864.v1/t8

    Values of $\Delta Z$ for the discovery sensitivity, as defined in the text, as a function of the invariant mass...

  • fig_03_jm

    Data from Figure 03

    10.17182/hepdata.144864.v1/t9

    Values of $\Delta Z$ for the discovery sensitivity, as defined in the text, as a function of the invariant mass...

  • fig_03_bm

    Data from Figure 03

    10.17182/hepdata.144864.v1/t10

    Values of $\Delta Z$ for the discovery sensitivity, as defined in the text, as a function of the invariant mass...

  • fig_03_bg

    Data from Figure 03

    10.17182/hepdata.144864.v1/t11

    Values of $\Delta Z$ for the discovery sensitivity, as defined in the text, as a function of the invariant mass...

  • fig_04

    Data from Figure 04

    10.17182/hepdata.144864.v1/t12

    The 95% CL upper limits on the cross section times acceptance ($A$), efficiency ($\epsilon$), and branching ratio ($B$) for Gaussian-shaped...

  • figaux_02

    Data from Auxiliary Figure 02

    10.17182/hepdata.144864.v1/t13

    Distributions of the anomaly score from the AE for data and five benchmark BSM models. Their legends, from top to...

  • figaux_03

    Data from Auxiliary Figure 03

    10.17182/hepdata.144864.v1/t14

    Invariant mass distributions of $j+Y$ for $M_{jY}$ > 0.3 TeV after preselection along with the fit from Eq.(1). The fit...

  • figaux_04

    Data from Auxiliary Figure 04

    10.17182/hepdata.144864.v1/t15

    Distributions of the anomaly score for data and several anomaly scenarios. The example BSM model (shown with the dashed blue...

Loading Data...

Ask a Question


Your question will be emailed to those involved with the submission. Please mention the relevant table.


Please log in to HEPData to send a question.