Robust Neural Particle Identification Models

The volume of data processed by the Large Hadron Collider experiments demands sophisticated selection rules typically based on machine learning algorithms. One of the shortcomings of these approaches is their profound sensitivity to the biases in training samples. In the case of particle identificat...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Temirkhanov, Aziz, Ryzhikov, Artem, Derkach, Denis, Hushchyn, Mikhail, Kazeev, Nikita, Mokhnenko, Sergei
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Temirkhanov, Aziz
Ryzhikov, Artem
Derkach, Denis
Hushchyn, Mikhail
Kazeev, Nikita
Mokhnenko, Sergei
description The volume of data processed by the Large Hadron Collider experiments demands sophisticated selection rules typically based on machine learning algorithms. One of the shortcomings of these approaches is their profound sensitivity to the biases in training samples. In the case of particle identification (PID), this might lead to degradation of the efficiency for some decays not present in the training dataset due to differences in input kinematic distributions. In this talk, we propose a method based on the Common Specific Decomposition that takes into account individual decays and possible misshapes in the training data by disentangling common and decay specific components of the input feature set. We show that the proposed approach reduces the rate of efficiency degradation for the PID algorithms for the decays reconstructed in the LHCb detector.
doi_str_mv 10.48550/arxiv.2212.07274
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2212_07274</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2212_07274</sourcerecordid><originalsourceid>FETCH-LOGICAL-a674-cf34ea85699593fab78d2177d3c2805736ec3ab83ad0f4d4b830df4265ad50cf3</originalsourceid><addsrcrecordid>eNotzskOgjAYBOBePBj0ATzJAwiWLrQcjXFL3GK4k58uSRMUU8Do24vLaeYykw-hSYJjJjnHc_BP94gJSUiMBRFsiGaXuuyaNjyazkMVnsG3TlUm3Glza511ClpX38JDrU3VjNDAQtWY8T8DlK9X-XIb7U-b3XKxjyAVLFKWMgOSp1nGM2qhFFKTRAhNFZGYC5oaRaGUFDS2TLO-YW0ZSTlojvt1gKa_2y-3uHt3Bf8qPuziy6ZvCnQ8jA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Robust Neural Particle Identification Models</title><source>arXiv.org</source><creator>Temirkhanov, Aziz ; Ryzhikov, Artem ; Derkach, Denis ; Hushchyn, Mikhail ; Kazeev, Nikita ; Mokhnenko, Sergei</creator><creatorcontrib>Temirkhanov, Aziz ; Ryzhikov, Artem ; Derkach, Denis ; Hushchyn, Mikhail ; Kazeev, Nikita ; Mokhnenko, Sergei</creatorcontrib><description>The volume of data processed by the Large Hadron Collider experiments demands sophisticated selection rules typically based on machine learning algorithms. One of the shortcomings of these approaches is their profound sensitivity to the biases in training samples. In the case of particle identification (PID), this might lead to degradation of the efficiency for some decays not present in the training dataset due to differences in input kinematic distributions. In this talk, we propose a method based on the Common Specific Decomposition that takes into account individual decays and possible misshapes in the training data by disentangling common and decay specific components of the input feature set. We show that the proposed approach reduces the rate of efficiency degradation for the PID algorithms for the decays reconstructed in the LHCb detector.</description><identifier>DOI: 10.48550/arxiv.2212.07274</identifier><language>eng</language><subject>Physics - High Energy Physics - Experiment ; Physics - Instrumentation and Detectors</subject><creationdate>2022-12</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2212.07274$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2212.07274$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Temirkhanov, Aziz</creatorcontrib><creatorcontrib>Ryzhikov, Artem</creatorcontrib><creatorcontrib>Derkach, Denis</creatorcontrib><creatorcontrib>Hushchyn, Mikhail</creatorcontrib><creatorcontrib>Kazeev, Nikita</creatorcontrib><creatorcontrib>Mokhnenko, Sergei</creatorcontrib><title>Robust Neural Particle Identification Models</title><description>The volume of data processed by the Large Hadron Collider experiments demands sophisticated selection rules typically based on machine learning algorithms. One of the shortcomings of these approaches is their profound sensitivity to the biases in training samples. In the case of particle identification (PID), this might lead to degradation of the efficiency for some decays not present in the training dataset due to differences in input kinematic distributions. In this talk, we propose a method based on the Common Specific Decomposition that takes into account individual decays and possible misshapes in the training data by disentangling common and decay specific components of the input feature set. We show that the proposed approach reduces the rate of efficiency degradation for the PID algorithms for the decays reconstructed in the LHCb detector.</description><subject>Physics - High Energy Physics - Experiment</subject><subject>Physics - Instrumentation and Detectors</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzskOgjAYBOBePBj0ATzJAwiWLrQcjXFL3GK4k58uSRMUU8Do24vLaeYykw-hSYJjJjnHc_BP94gJSUiMBRFsiGaXuuyaNjyazkMVnsG3TlUm3Glza511ClpX38JDrU3VjNDAQtWY8T8DlK9X-XIb7U-b3XKxjyAVLFKWMgOSp1nGM2qhFFKTRAhNFZGYC5oaRaGUFDS2TLO-YW0ZSTlojvt1gKa_2y-3uHt3Bf8qPuziy6ZvCnQ8jA</recordid><startdate>20221214</startdate><enddate>20221214</enddate><creator>Temirkhanov, Aziz</creator><creator>Ryzhikov, Artem</creator><creator>Derkach, Denis</creator><creator>Hushchyn, Mikhail</creator><creator>Kazeev, Nikita</creator><creator>Mokhnenko, Sergei</creator><scope>GOX</scope></search><sort><creationdate>20221214</creationdate><title>Robust Neural Particle Identification Models</title><author>Temirkhanov, Aziz ; Ryzhikov, Artem ; Derkach, Denis ; Hushchyn, Mikhail ; Kazeev, Nikita ; Mokhnenko, Sergei</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a674-cf34ea85699593fab78d2177d3c2805736ec3ab83ad0f4d4b830df4265ad50cf3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Physics - High Energy Physics - Experiment</topic><topic>Physics - Instrumentation and Detectors</topic><toplevel>online_resources</toplevel><creatorcontrib>Temirkhanov, Aziz</creatorcontrib><creatorcontrib>Ryzhikov, Artem</creatorcontrib><creatorcontrib>Derkach, Denis</creatorcontrib><creatorcontrib>Hushchyn, Mikhail</creatorcontrib><creatorcontrib>Kazeev, Nikita</creatorcontrib><creatorcontrib>Mokhnenko, Sergei</creatorcontrib><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Temirkhanov, Aziz</au><au>Ryzhikov, Artem</au><au>Derkach, Denis</au><au>Hushchyn, Mikhail</au><au>Kazeev, Nikita</au><au>Mokhnenko, Sergei</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Robust Neural Particle Identification Models</atitle><date>2022-12-14</date><risdate>2022</risdate><abstract>The volume of data processed by the Large Hadron Collider experiments demands sophisticated selection rules typically based on machine learning algorithms. One of the shortcomings of these approaches is their profound sensitivity to the biases in training samples. In the case of particle identification (PID), this might lead to degradation of the efficiency for some decays not present in the training dataset due to differences in input kinematic distributions. In this talk, we propose a method based on the Common Specific Decomposition that takes into account individual decays and possible misshapes in the training data by disentangling common and decay specific components of the input feature set. We show that the proposed approach reduces the rate of efficiency degradation for the PID algorithms for the decays reconstructed in the LHCb detector.</abstract><doi>10.48550/arxiv.2212.07274</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2212.07274
ispartof
issn
language eng
recordid cdi_arxiv_primary_2212_07274
source arXiv.org
subjects Physics - High Energy Physics - Experiment
Physics - Instrumentation and Detectors
title Robust Neural Particle Identification Models
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-11T09%3A27%3A32IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Robust%20Neural%20Particle%20Identification%20Models&rft.au=Temirkhanov,%20Aziz&rft.date=2022-12-14&rft_id=info:doi/10.48550/arxiv.2212.07274&rft_dat=%3Carxiv_GOX%3E2212_07274%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true