[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-126573-105":59,"doc-detail-126573-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","assessment-of-few-hits-machine-learning-classification-algorithms-for-low-energy-physics-in-liquid-argon-detectors","Assessment of Few-Hits Machine Learning Classification Algorithms for Low Energy Physics in Liquid Argon Detectors","","Few-hits events in low-energy liquid-argon TPCs contain information that conventional deterministic classification methods struggle to fully exploit. The study evaluates machine-learning approaches against deterministic algorithms for challenging single-versus double-beta event classification. Convolutional Neural Networks and TransformerEncoder models show superior performance, with a direct comparison of their strengths and limitations. The work applies these methods to optimize detector parameters, focusing on DUNE Phase II liquid-argon detectors.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/assessment-of-few-hits-machine-learning-classification-algorithms-for-low-energy-physics-in-liquid-argon-detectors/126573/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/assessment-of-few-hits-machine-learning-classification-algorithms-for-low-energy-physics-in-liquid-argon-detectors/126573.png","ImageObject",300,407,{"name":92,"@type":93},"Himbo","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-17","2026-08-05",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",15,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"Why are conventional deterministic algorithms less effective for low-energy few-hits events in liquid argon detectors?","Question",{"text":113,"@type":114},"Few-meV events produce only a small number of hits, and the resulting sparse information is difficult for conventional deterministic classifiers to exploit effectively.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"Which machine learning models outperform deterministic algorithms in the evaluated single-versus double-beta classification problem?",{"text":118,"@type":114},"Convolutional Neural Networks (CNN) and TransformerEncoder methods both outperform deterministic algorithms in this challenging low-energy classification task.",{"name":120,"@type":111,"acceptedAnswer":121},"How are the ML methods used to support detector development for DUNE Phase II?",{"text":122,"@type":114},"The study uses CNN and TransformerEncoder approaches to optimize detector parameters, emphasizing design choices relevant to DUNE Phase II and the Module of Opportunity R&D direction.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},126573,1785933403,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":130,"read_time":41,"preview_page_count":4},687207017582,"https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d","arXiv :2305 .09744v1 [physics .ins-det] 16 May 2023  \nEur. Phys. J. C manuscript No.  \n(will be inserted by the editor)  \nAssessment of few-hits machine learning classi􀀌cation algorithms for low energy physics in liquid argon detectors  \nM. Biassoni 1 , A. Giachero 1 ,2 , M. Grossi3 , D. Gu􀀋anti 1 ,2 , D. Labranca 1 ,2 ,  \nR. Moretti 1 ,2 ,a , M. Rossi3 ,4 ,5 ,b , F. Terranova 1 ,2 , S. Vallecorsa3  \n1 Istituto Nazionale di Fisica Nucleare (INFN), Sezione di Milano-Bicocca, 20126 Milan, Italy  \n2 Dipartimento di Fisica \"G. Occhialini\", Universit􀀒a di Milano-Bicocca, 20126 Milan, Italy  \n3 European Organization for Nuclear Research (CERN), Geneva CH-1211, Switzerland  \n4 Dipartimento di Fisica, Universit􀀒a degli Studi di Milano, 20133 Milan, Italy  \n5 Istituto Nazionale di Fisica Nucleare (INFN) Sezione di Milano, 20133 Milan, Italy  \nReceived: date / Accepted: date  \nAbstract The physics potential of massive liquid argon TPCs in the low-energy regime is still to be fully reaped because few-hits events encode information that can hardly be exploited by conventional classi􀀌cation algorithms. Machine learning (ML) techniques give their best in these types of classi􀀌cation problems. In this paper, we evaluate their performance against conventional (deterministic) algorithms. We demonstrate that both Convolutional Neural Networks (CNN) and TransformerEncoder methods outperform deterministic algorithms in one of the most challenging classi􀀌cation problems of low-energy physics (single-versus double-beta events) . We discuss the advantages and pitfalls of TransformerEncoder methods versus CNN and employ these methods to optimize the detector parameters, with an emphasis on the DUNE Phase II detectors (\\Module of Opportunity\") .  \nKeywords liquid argon TPC 􀀁 machine learning 􀀁 particle identi􀀌cation 􀀁 Transformer-Encoder  \n1 Introduction  \nLiquid argon detectors play a prominent role in neutrino physics and-thanks to experiments like DUNE [1, 2] and DarkSide [3, 4] -will be one of the technologies of choice for the next generation of accelerator neutrino experiments and direct search of dark matter. Such prominence is grounded on scalability. DUNE, in particular, has brought the liquid argon TPC technique (LArTPC) to an unprecedented scale after the development of cryostats that do not need to be evacuated to reach the purity needed in TPCs (membrane  \nae-mail: [roberto.moretti@mib.infn.it](roberto.moretti@mib.infn.it)  \n[b](be-mail: marco.rossi@cern.ch)[e-mail: marco.rossi@cern.ch](be-mail: marco.rossi@cern.ch)  \ncryostat) [5, 6 , 7] . Similarly, DarkSide is commissioning high-throughput facilities for depleted underground argon extraction, puri􀀌cation, and distillation [8, 9] .  \nIn the last few years and, notably, in the course of the 2021 Snowmass process, several collaborations have formed to 􀀌ll the gap between LarTPC for beam neutrinos and detectors for rare event searches [10, 11 , 12 , 13 , 14] . The 􀀌rst and second DUNE modules were engineered to achieve maximum performance for the observation of GeV-scale neutrinos but a wealth of DUNE physics resides in the observation of MeV events. Some of these channels are being actively pursued by DUNE because the event threshold E is located at E > 10 MeV. They are supernova neutrinos, sub-GeV atmospheric neutrinos, and the neutrinos originating from the sun from helium-proton fusion (\"hep neutrinos\")[15, 16] . Other channels are currently outside the DUNEscope but may be addressed by the DUNE Phase II modules (third and fourth modules) . Recent studies conducted in the framework of the Module of Opportunity (MoO) R&D e􀀋ort [17] indicate that a module with improved radiopurity, light collection e􀀎ciency, and granularity can access a rich physics portfolio: enhance the sensitivity to solar neutrinos produced by the 8 B branch of the pp-chain, perform neutrinoless double beta decay searches, boosted dark matter, and the direct detection of WIMP-like dark matter candidates [13, 18 , 19]","cbCaiprQyxaE9uoa","https://ap.wps.com/l/cbCaiprQyxaE9uoa","pdf",673946,12,"English","# Abstract\n# Introduction\n## Low-energy physics goals for liquid argon detectors\n## Few-hits event challenge and scalability\n# Machine learning approach\n## CNN and TransformerEncoder methods\n## Comparison with deterministic algorithms\n# Detector optimization for DUNE Phase II\n## Module of Opportunity design considerations\n# Conclusions","[{\"question\":\"Why are conventional deterministic algorithms less effective for low-energy few-hits events in liquid argon detectors?\",\"answer\":\"Few-meV events produce only a small number of hits, and the resulting sparse information is difficult for conventional deterministic classifiers to exploit effectively.\"},{\"question\":\"Which machine learning models outperform deterministic algorithms in the evaluated single-versus double-beta classification problem?\",\"answer\":\"Convolutional Neural Networks (CNN) and TransformerEncoder methods both outperform deterministic algorithms in this challenging low-energy classification task.\"},{\"question\":\"How are the ML methods used to support detector development for DUNE Phase II?\",\"answer\":\"The study uses CNN and TransformerEncoder approaches to optimize detector parameters, emphasizing design choices relevant to DUNE Phase II and the Module of Opportunity R\\u0026D direction.\"}]","Assessment of Few-Hits Machine Learning Classification Algorithms for Low Energy Physics in Liquid Argon Detectors | PDF"]