[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123589-en":3,"doc-seo-123589-105":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":20,"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},123589,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","Probing Heavy Neutrinos at the LHC from Fat-jet using Machine Learning - Abstract","The study investigates using machine learning to search for heavy neutrinos at hadron colliders through hadronic final states containing a fat-jet signature, within processes pp → W* → ℓ N and N → ℓ W*, where the W decays to jets. Gradient Boosted Decision Trees and Multi-Layer Perceptrons are trained on multivariate observables incorporating jet substructure at √s = 13, 27, and 100 TeV. Invariant masses of variable systems and lepton-derived observables provide the strongest discrimination against Standard Model backgrounds. Machine learning improves limits on active-sterile mixing by roughly one order of magnitude versus cut-based analyses, reaching V2ℓN ≲ 10^-4 for mN between 100 GeV and 1 TeV.","arXiv :2303 . 15920v1 [hep-ph] 28 Mar 2023  \nProbing Heavy Neutrinos at the LHC from Fat-jet using Machine Learning  \nWei Liu,a Jing Li, b Zixiang Chen,a Hao Sunb;1  \na Department of Applied Physics and MIIT Key Laboratory of Semiconductor Microstructure and Quantum Sensing, Nanjing University of Science and Technology, Nanjing 210094, China  \nb Institute of Theoretical Physics, School of Physics, Dalian University of Technology, No. 2 Linggong Road, Dalian, Liaoning, 116024, P.R. China  \nE-mail: [wei.liu@njust.edu.cn](wei.liu@njust.edu.cn), [haosun@dlut.edu.cn](haosun@dlut.edu.cn)  \nAbstract: We explore the potential to use machine learning methods to search for heavy neutrinos, from their hadronic final states including a fat-jet signal, via the processes pp ! W􀀆􀀃 ! 􀀖􀀆 N ! 􀀖􀀆 􀀖􀀇 W􀀆 ! 􀀖􀀆 􀀖􀀇 J at hadron colliders. We use either the Gradient Boosted Decision Tree or Multi-Layer Perceptron methods to analyse the observables incorporating the jet substructure information, which is performed at hadron colliders with ps = 13, 27, 100 TeV. It is found that, among the observables, the invariant masses of variable system and the observables from the leptons are the most powerful ones to distinguish the signal from the background. With the help of machine learning techniques, the limits on the active-sterile mixing have been improved by about one magnitude comparing to the cut-based analyses, with V2􀀖N . 10􀀀4 for the heavy neutrinos with masses, 100 GeV \u003C mN \u003C 1 TeV.  \n1 Corresponding author.  \nContents  \n1 Introduction 1  \n2 Inverse Seesaw Model 2  \n3 Analysis 4  \n3.1 Cut Based Analysis 5  \n3.2 Machine Learning 6  \n4 Sensitivity 10  \n5 Conclusion 13  \n1 Introduction  \nThe observation of tiny neutrino masses is one of the most direct evidence for the existence of the new physics beyond the Standard Model (SM) . It can be explained by the canonical type-I seesaw, but the Dirac Yukawa couplings required by the eV scale active neutrino masses are not natural, YD 􀀘 10􀀀6, if the right-handed (heavy) neutrinos are in electroweak (EW) scale [1–4] . This results in the suppression on the active-sterile mixing (VlN), making the heavy neutrinos long-lived, which received a lot of attention [5–9] . Such suppression can be lifted away, if the inverse seesaw is considered where the smallness of the neutrino masses can be absorbed by a naturally small lepton number violating parameter, while additional SM singlet fermions S are also introduced [10 , 11] . Thus, the Yukawa couplings can be of order one, and the mixing between the active and heavy neutrinos VlN can be sizeable, with Vl2N . 10􀀀3 from electroweak precision data (EWPD) [12–15], leading to very rich phenomenology, especially at colliders [16–49] .  \nThe most considered production channel of the heavy neutrinos (N), are the chargedcurrent Drell-Yan process pp ! W􀀆(􀀃) ! l􀀆 N, with the subsequent decay N ! l􀀇 W􀀆 , and the W dominantly decays into dijets. In the type-I seesaw, one can be benefited from the Majorana nature of the heavy neutrinos, leading to the \"smoking-gun\" signal of samesign dilepton, l􀀆 l􀀆 , which is considered to be almost SM background-free [4] . However, the heavy neutrinos and singlet S are mass degenerate in the inverse seesaw, forming a pseusoDirac fermion pairs, while the Majorana component are highly suppressed. Therefore, we rely on the invariant masses of the lepton and dijets final states to search for the heavy neutrinos, which can be severely contaminated by the SM processes.  \nThis motivates the introduction of new techniques, such as the jet substructure techniques, which can be a powerful tool to help to resolve the problems. As considered in Ref. [50–52], the heavy neutrinos can lead to collimated topology of the dijets, forming a  \nso-called \"fat-jet\" system. This can be of great help to improve significance and mitigating background. More recently, machine learning (ML) techniques are also employed in Ref. [52 , 53] to improve the sensitivity on the active-st","cbCaiaTlUqDrx5FF","https://ap.wps.com/l/cbCaiaTlUqDrx5FF","pdf",1385382,1,19,"English","en",105,"# Introduction\n## Heavy neutrino motivation and inverse seesaw framework\n## Production and search strategy using fat-jet observables\n# Inverse Seesaw Model\n# Analysis\n## Cut Based Analysis\n## Machine Learning\n# Sensitivity\n# Conclusion","[{\"question\":\"What heavy-neutrino production and decay channel is used in the analysis?\",\"answer\":\"The approach considers pp collisions producing a W* that leads to ℓN, followed by N decaying to ℓW*, with the W* dominantly decaying into dijets that form a fat-jet signature.\"},{\"question\":\"Which machine learning methods are employed to distinguish signal from background?\",\"answer\":\"The document uses Gradient Boosted Decision Tree (GBDT) and Multi-Layer Perceptron (MLP) models implemented in the scikit-learn framework, trained on observables including jet substructure information.\"},{\"question\":\"Which observables provide the strongest discrimination power?\",\"answer\":\"The invariant masses of the variable system and observables derived from leptons are identified as the most powerful to separate the heavy-neutrino signal from Standard Model backgrounds.\"}]","Probing Heavy Neutrinos at the LHC from Fat-jet using Machine Learning - Abstract | PDF",1785817512,48,{"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},"probing-heavy-neutrinos-at-the-lhc-from-fat-jet-using-machine-learning-abstract","",{"@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/probing-heavy-neutrinos-at-the-lhc-from-fat-jet-using-machine-learning-abstract/123589/",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":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What heavy-neutrino production and decay channel is used in the analysis?","Question",{"text":75,"@type":76},"The approach considers pp collisions producing a W* that leads to ℓN, followed by N decaying to ℓW*, with the W* dominantly decaying into dijets that form a fat-jet signature.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which machine learning methods are employed to distinguish signal from background?",{"text":80,"@type":76},"The document uses Gradient Boosted Decision Tree (GBDT) and Multi-Layer Perceptron (MLP) models implemented in the scikit-learn framework, trained on observables including jet substructure information.",{"name":82,"@type":73,"acceptedAnswer":83},"Which observables provide the strongest discrimination power?",{"text":84,"@type":76},"The invariant masses of the variable system and observables derived from leptons are identified as the most powerful to separate the heavy-neutrino signal from Standard Model backgrounds.","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},[93,97,101,105,110,115,120,123,128,131,135],{"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":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},"General","general"]