skip to main content


Title: Pile-up mitigation using attention
Abstract

Particle production from secondary proton-proton collisions, commonly referred to as pile-up, impair the sensitivity of both new physics searches and precision measurements at large hadron collider (LHC) experiments. We propose a novel algorithm,Puma, for modeling pile-up with the help of deep neural networks based on sparse transformers. These attention mechanisms were developed for natural language processing but have become popular in other applications. In a realistic detector simulation, our method outperforms classical benchmark algorithms for pile-up mitigation in key observables. It provides a perspective for mitigating the effects of pile-up in the high luminosity era of the LHC, where up to 200 proton-proton collisions are expected to occur simultaneously.

 
more » « less
NSF-PAR ID:
10371378
Author(s) / Creator(s):
; ; ; ; ;
Publisher / Repository:
IOP Publishing
Date Published:
Journal Name:
Machine Learning: Science and Technology
Volume:
3
Issue:
2
ISSN:
2632-2153
Page Range / eLocation ID:
Article No. 025012
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. A<sc>bstract</sc>

    A search for supersymmetry is presented in events with a single charged lepton, electron or muon, and multiple hadronic jets. The data correspond to an integrated luminosity of 138 fb1of proton-proton collisions at a center-of-mass energy of 13 TeV, recorded by the CMS experiment at the CERN LHC. The search targets gluino pair production, where the gluinos decay into final states with the lightest supersymmetric particle (LSP) and either a top quark-antiquark ($$ \textrm{t}\overline{\textrm{t}} $$tt¯) pair, or a light-flavor quark-antiquark ($$ \textrm{q}\overline{\textrm{q}} $$qq¯) pair and a virtual or on-shell W boson. The main backgrounds,$$ \textrm{t}\overline{\textrm{t}} $$tt¯pair and W+jets production, are suppressed by requirements on the azimuthal angle between the momenta of the lepton and of its reconstructed parent W boson candidate, and by top quark and W boson identification based on a machine-learning technique. The number of observed events is consistent with the expectations from standard model processes. Limits are evaluated on supersymmetric particle masses in the context of two simplified models of gluino pair production. Exclusions for gluino masses reach up to 2120 (2050) GeV at 95% confidence level for a model with gluino decay to a$$ \textrm{t}\overline{\textrm{t}} $$tt¯pair (a$$ \textrm{q}\overline{\textrm{q}} $$qq¯pair and a W boson) and the LSP. For the same models, limits on the mass of the LSP reach up to 1250 (1070) GeV.

     
    more » « less
  2. A<sc>bstract</sc>

    The collective behavior of$$ {\textrm{K}}_{\textrm{S}}^0 $$KS0and$$ \Lambda /\overline{\Lambda} $$Λ/Λ¯strange hadrons is studied by measuring the elliptic azimuthal anisotropy (v2) using the scalar-product and multiparticle correlation methods. Proton-lead (pPb) collisions at a nucleon-nucleon center-of-mass energy$$ \sqrt{s_{\textrm{NN}}} $$sNN= 8.16 TeV and lead-lead (PbPb) collisions at$$ \sqrt{s_{\textrm{NN}}} $$sNN= 5.02 TeV collected by the CMS experiment at the LHC are investigated. Nonflow effects in the pPb collisions are studied by using a subevent cumulant analysis and by excluding events where a jet with transverse momentum greater than 20 GeV is present. The strange hadronv2values extracted in pPb collisions via the four- and six-particle correlation method are found to be nearly identical, suggesting the collective behavior. Comparisons of the pPb and PbPb results for both strange hadrons and charged particles illustrate how event-by-event flow fluctuations depend on the system size.

     
    more » « less
  3. A<sc>bstract</sc>

    The production of inclusive, prompt and non-prompt J/ψwas studied for the first time at midrapidity (−1.37< ycms<0.43) in p-Pb collisions at$$ \sqrt{s_{\textrm{NN}}} $$sNN= 8.16 TeV with the ALICE detector at the LHC. The inclusive J/ψmesons were reconstructed in the dielectron decay channel in the transverse momentum (pT) interval 0< pT<14 GeV/cand the prompt and non-prompt contributions were separated on a statistical basis forpT>2 GeV/c. The study of the J/ψmesons in the dielectron channel used for the first time in ALICE online single-electron triggers from the Transition Radiation Detector, providing a data sample corresponding to an integrated luminosity of 689 ± 13μb1. The proton-proton reference cross section for inclusive J/ψwas obtained based on interpolations of measured data at different centre-of-mass energies and a universal function describing thepT-differential J/ψproduction cross sections. ThepT-differential nuclear modification factorsRpPbof inclusive, prompt, and non-prompt J/ψare consistent with unity and described by theoretical models implementing only nuclear shadowing.

     
    more » « less
  4. A<sc>bstract</sc>

    Three searches are presented for signatures of physics beyond the standard model (SM) inττfinal states in proton-proton collisions at the LHC, using a data sample collected with the CMS detector at$$ \sqrt{s} $$s= 13 TeV, corresponding to an integrated luminosity of 138 fb1. Upper limits at 95% confidence level (CL) are set on the products of the branching fraction for the decay intoτleptons and the cross sections for the production of a new bosonϕ, in addition to the H(125) boson, via gluon fusion (ggϕ) or in association with b quarks, ranging from$$ \mathcal{O} $$O(10 pb) for a mass of 60 GeV to 0.3 fb for a mass of 3.5 TeV each. The data reveal two excesses for ggϕproduction with localp-values equivalent to about three standard deviations atmϕ= 0.1 and 1.2 TeV. In a search fort-channel exchange of a vector leptoquark U1, 95% CL upper limits are set on the dimensionless U1leptoquark coupling to quarks andτleptons ranging from 1 for a mass of 1 TeV to 6 for a mass of 5 TeV, depending on the scenario. In the interpretations of the$$ {M}_{\textrm{h}}^{125} $$Mh125and$$ {M}_{\textrm{h},\textrm{EFT}}^{125} $$Mh,EFT125minimal supersymmetric SM benchmark scenarios, additional Higgs bosons with masses below 350 GeV are excluded at 95% CL.

     
    more » « less
  5. A<sc>bstract</sc>

    A search for pairs of dijet resonances with the same mass is conducted in final states with at least four jets. Results are presented separately for the case where the four jet production proceeds via an intermediate resonant state and for nonresonant production. The search uses a data sample corresponding to an integrated luminosity of 138 fb1collected by the CMS detector in proton-proton collisions at$$ \sqrt{s} $$s= 13 TeV. Model-independent limits, at 95% confidence level, are reported on the production cross section of four-jet and dijet resonances. These first LHC limits on resonant pair production of dijet resonances via high mass intermediate states are applied to a signal model of diquarks that decay into pairs of vector-like quarks, excluding diquark masses below 7.6 TeV for a particular model scenario. There are two events in the tails of the distributions, each with a four-jet mass of 8 TeV and an average dijet mass of 2 TeV, resulting in local and global significances of 3.9 and 1.6 standard deviations, respectively, if interpreted as a signal. The nonresonant search excludes pair production of top squarks with masses between 0.50 TeV to 0.77 TeV, with the exception of a small interval between 0.52 and 0.58 TeV, for supersymmetricR-parity-violating decays to quark pairs, significantly extending previous limits. Here, the most significant excess above the predicted background occurs at an average dijet mass of 0.95 TeV, for which the local and global significances are 3.6 and 2.5 standard deviations, respectively.

     
    more » « less