Cargando…

Particle identification with machine learning in ALICE Run 3

The main focus of the ALICE experiment, quark--gluon plasma measurements, requires accurate particle identification (PID). The ALICE subdetectors allow identifying particles over a broad momentum interval ranging from about 100 MeV/c up to 20 GeV/c. However, a machine learning (ML) model can explore...

Descripción completa

Detalles Bibliográficos
Autores principales: Karwowska, Maja, Jakubowska, Monika, Graczykowski, Łukasz, Deja, Kamil, Kasak, Miłosz
Lenguaje:eng
Publicado: 2023
Materias:
Acceso en línea:http://cds.cern.ch/record/2871452
_version_ 1780978535985315840
author Karwowska, Maja
Jakubowska, Monika
Graczykowski, Łukasz
Deja, Kamil
Kasak, Miłosz
author_facet Karwowska, Maja
Jakubowska, Monika
Graczykowski, Łukasz
Deja, Kamil
Kasak, Miłosz
author_sort Karwowska, Maja
collection CERN
description The main focus of the ALICE experiment, quark--gluon plasma measurements, requires accurate particle identification (PID). The ALICE subdetectors allow identifying particles over a broad momentum interval ranging from about 100 MeV/c up to 20 GeV/c. However, a machine learning (ML) model can explore more detector information. During LHC Run 2, preliminary studies with Random Forests obtained much higher efficiencies and purities for selected particles than standard techniques. For Run 3, we investigate Domain Adaptation Neural Networks that account for the discrepancies between the Monte Carlo simulations and the experimental data. Preliminary studies show that domain adaptation improves particle classification. Moreover, the solution is extended with Feature Set Embedding and attention to give the network more flexibility to train on data with various sets of detector signals. PID ML is already integrated with the ALICE Run 3 Analysis Framework. Preliminary results for the PID of selected particle species, including real-world analyzes, are discussed as well as the possible optimizations.
id cern-2871452
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2023
record_format invenio
spelling cern-28714522023-10-03T15:52:06Zhttp://cds.cern.ch/record/2871452engKarwowska, MajaJakubowska, MonikaGraczykowski, ŁukaszDeja, KamilKasak, MiłoszParticle identification with machine learning in ALICE Run 3physics.ins-detDetectors and Experimental Techniquesnucl-exNuclear Physics - Experimenthep-exParticle Physics - ExperimentThe main focus of the ALICE experiment, quark--gluon plasma measurements, requires accurate particle identification (PID). The ALICE subdetectors allow identifying particles over a broad momentum interval ranging from about 100 MeV/c up to 20 GeV/c. However, a machine learning (ML) model can explore more detector information. During LHC Run 2, preliminary studies with Random Forests obtained much higher efficiencies and purities for selected particles than standard techniques. For Run 3, we investigate Domain Adaptation Neural Networks that account for the discrepancies between the Monte Carlo simulations and the experimental data. Preliminary studies show that domain adaptation improves particle classification. Moreover, the solution is extended with Feature Set Embedding and attention to give the network more flexibility to train on data with various sets of detector signals. PID ML is already integrated with the ALICE Run 3 Analysis Framework. Preliminary results for the PID of selected particle species, including real-world analyzes, are discussed as well as the possible optimizations.arXiv:2309.07768oai:cds.cern.ch:28714522023-09-14
spellingShingle physics.ins-det
Detectors and Experimental Techniques
nucl-ex
Nuclear Physics - Experiment
hep-ex
Particle Physics - Experiment
Karwowska, Maja
Jakubowska, Monika
Graczykowski, Łukasz
Deja, Kamil
Kasak, Miłosz
Particle identification with machine learning in ALICE Run 3
title Particle identification with machine learning in ALICE Run 3
title_full Particle identification with machine learning in ALICE Run 3
title_fullStr Particle identification with machine learning in ALICE Run 3
title_full_unstemmed Particle identification with machine learning in ALICE Run 3
title_short Particle identification with machine learning in ALICE Run 3
title_sort particle identification with machine learning in alice run 3
topic physics.ins-det
Detectors and Experimental Techniques
nucl-ex
Nuclear Physics - Experiment
hep-ex
Particle Physics - Experiment
url http://cds.cern.ch/record/2871452
work_keys_str_mv AT karwowskamaja particleidentificationwithmachinelearninginalicerun3
AT jakubowskamonika particleidentificationwithmachinelearninginalicerun3
AT graczykowskiłukasz particleidentificationwithmachinelearninginalicerun3
AT dejakamil particleidentificationwithmachinelearninginalicerun3
AT kasakmiłosz particleidentificationwithmachinelearninginalicerun3