[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123890-en":3,"doc-seo-123890-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},123890,137441390410,"Hazel","https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984",8,"Research & Report","Interpretable Machine Learning Methods Applied to Jet Background Subtraction in Heavy Ion Collisions","Jet measurements in heavy ion collisions help constrain quark gluon plasma properties, but fluctuating soft-particle backgrounds restrict kinematic reach and limit precision of background subtraction. The work applies symbolic regression to extract an interpretable functional representation of a deep neural network trained to subtract jet backgrounds. The extracted representation is shown to be approximately equivalent to a particle multiplicity-based method, supporting that interpretable machine learning can reveal underlying physical mechanisms while clarifying when improvements arise.","[https://doi.org/10.1103/PhysRevC.108.L021901](https://doi.org/10.1103/PhysRevC.108.L021901)  \nInterpretable Machine Learning Methods Applied to Jet Background Subtraction in Heavy Ion Collisions  \narXiv :2303 .08275v2 [hep-ex] 23 Aug 2023  \nTanner Mengel, Patrick Steffanic, Charles Hughes, Antonio Carlos Oliveira da Silva, and Christine Nattrass  \nUniversity of Tennessee, Knoxville, TN, USA-37996 .  \n(Dated: August 24, 2023)  \nJet measurements in heavy ion collisions can provide constraints on the properties of the quark gluon plasma, but the kinematic reach is limited by a large, fluctuating background. We present a novel application of symbolic regression to extract a functional representation of a deep neural network trained to subtract background from jets in heavy ion collisions. We show that the deep neural network is approximately the same as a method using the particle multiplicity in a jet. This demonstrates that interpretable machine learning methods can provide insight into underlying physical processes.  \nI. INTRODUCTION  \nThe Quark Gluon Plasma (QGP) is a hot, dense, strongly interacting liquid of quarks and gluons that is created briefly in high energy heavy ion collisions [1–4] . Measurements of jets produced by hard scatterings between partons in heavy ion collisions can be used to investigate the properties of the QGP [5] . Quantitative comparisons between jet measurements and physics models can provide further constraints on these properties [6, 7] . However, heavy ion events are dominated by a fluctuating background of soft particles not due to hard scatterings. The details of these fluctuations are sensitive to correlations from hydrodynamical flow and the shape of the single particle spectra [8], and as such are unlikely to be exactly the same in data and models. Mixed events are able to successfully describe the background in measurements of hadron-jet correlations by the STAR collaboration [9] at the Relativistic Heavy Ion Collider (RHIC) . Studies of the background at the Large Hadron Collider (LHC) by the ALICE Collaboration found that the distribution of background energy density in random cones is well described by a random background with correlations due to hydrodynamical flow and Poissonian fluctuations [10] . A better understanding of this background will facilitate more precise jet measurements for comparisons between data and models.  \nMeasurement precision and kinematic range is limited by the ability to correct for this background and its fluctuations. Background correction in jet measurements requires subtraction of contributions from soft particles within the jet, and suppression of fluctuations which have been reconstructed as combinatorial jets. At low momenta, combinatorial jets limit the kinematic reach of the measurement. Improved background subtraction methods would increase measurements’ sensitivity to partonic energy loss. Measurements of jet spectra which extend to low momenta primarily use the area method [11] for background subtraction. This method was initially proposed to correct for the underlying event in p+p collisionsin high pile-up conditions [11] and has also been applied to heavy ion collisions [12–15] .  \nThe complexity of jet background subtraction makes it an interesting environment to apply machine learning techniques. However, application of machine learning methods to background subtraction should be handled with care since models are not able to fully reproduce background fluctuations in heavy ion collisions [8] . Nuclear physics has prioritized the continued advancement in machine learning analysis techniques with a focus on interpretable methods that are robust, provide clear uncertainty quantification, and are explainable [16] . Applications of non-interpretable machine learning methods are insufficient when models available for training maybe inaccurate, when it may be necessary to understand the method to interpret the results, or when a result is needed out","cbCaislU4ksFF6bP","https://ap.wps.com/l/cbCaislU4ksFF6bP","pdf",687961,1,7,"English","en",105,"# Introduction\n## Quark gluon plasma and the role of jet measurements\n## Background fluctuations and existing subtraction approaches\n## Limits of non-interpretable machine learning\n# Method\n## Simulation\n## Neural-network training and symbolic regression","[{\"question\":\"Why is jet background subtraction critical in heavy ion collisions?\",\"answer\":\"Soft-particle fluctuations dominate heavy ion events, and uncertainties from these fluctuations limit the precision and kinematic range of jet measurements. Subtraction methods must remove soft contributions and suppress reconstructed combinatorial-jet fluctuations, especially at low momenta.\"},{\"question\":\"How does symbolic regression contribute to interpretability in this study?\",\"answer\":\"Symbolic regression is used to derive a functional representation that maps what the deep neural network has learned for background subtraction. This enables comparison to simpler physically motivated forms rather than relying only on an opaque model.\"},{\"question\":\"What relationship is found between the neural-network method and a multiplicity-based approach?\",\"answer\":\"The symbolic-regression-derived functional description of the deep neural network is approximately the same as a method using particle multiplicity in a jet. This suggests the network improvement corresponds to a physically interpretable signal captured by the multiplicity method.\"}]","Interpretable Machine Learning Methods Applied to Jet Background Subtraction in Heavy Ion Collisions | PDF",1785819091,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"interpretable-machine-learning-methods-applied-to-jet-background-subtraction-in-heavy-ion-collisions","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/interpretable-machine-learning-methods-applied-to-jet-background-subtraction-in-heavy-ion-collisions/123890/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is jet background subtraction critical in heavy ion collisions?","Question",{"text":75,"@type":76},"Soft-particle fluctuations dominate heavy ion events, and uncertainties from these fluctuations limit the precision and kinematic range of jet measurements. Subtraction methods must remove soft contributions and suppress reconstructed combinatorial-jet fluctuations, especially at low momenta.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does symbolic regression contribute to interpretability in this study?",{"text":80,"@type":76},"Symbolic regression is used to derive a functional representation that maps what the deep neural network has learned for background subtraction. This enables comparison to simpler physically motivated forms rather than relying only on an opaque model.",{"name":82,"@type":73,"acceptedAnswer":83},"What relationship is found between the neural-network method and a multiplicity-based approach?",{"text":84,"@type":76},"The symbolic-regression-derived functional description of the deep neural network is approximately the same as a method using particle multiplicity in a jet. This suggests the network improvement corresponds to a physically interpretable signal captured by the multiplicity method.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]