[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127468-en":3,"doc-seo-127468-105":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},127468,962084925290,"Ophelia","https://ap-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","Machine Learning Discoveries and Scientific Understanding in Particle Physics - Problems and Prospects","Particle physicists adopt machine learning methods early, and modern ML systems now support tasks ranging from detector signal reconstruction to event simulation and statistical inference in final analyses. Evidence suggests deep neural networks may acquire complex physical concepts and help discover new particles or phenomena independently. The chapter examines the epistemic challenges posed by discovery without human understanding, and analyzes concrete and prospective ways to mitigate these challenges for scientific progress and explanation.","Machine Learning Discoveries and Scienti✜c Understanding in Particle Physics: Problems  \nand Prospects * Florian J. Boge❸❹ & Henk W. de Regt➓  \nParticle physicists have been among the early adopters of Machine Learning (ML) methods, the most notable ML systems being Deep Neural Networks (DNNs) . Today, ML✬s use in Particle Physics (PP) ranges from the reconstruction of signals inside the detector to the simulation of events and the determination of statistical ratios in the ✜nal analysis. Most intriguingly, there is some evidence which suggests that DNNs might be able to independently acquire complex physical concepts✖concepts that are relevant for the discovery and understanding of new particles and phenomena. We here argue that these two possibilities, that of discovering novel concepts per se, and that of discovering novel phenomena by means of them, pose epistemic challenges for particle physicists. In turn, we will analyse ways of mitigating these challenges, both actual and at present merely possible.  \nKeywords particle physics • scienti✜c discovery • concepts • phenomena • understanding • deep learning  \n1 Introduction  \nParticle physicists face a tremendous amount of data in their experiments. Inside detectors at the Large Hadron Collider (LHC) at CERN in Geneva, numbers can ramp up to 1 billion collisions per second, leading to an incredible amount of some petabyte of collision-data per second.1 These data are ✜ltered down by sophisticated ❵trigger systems✬ to just some hundreds of megabytes per second, but of course this still means a huge amount of data to be stored and analysed. Furthermore, the response of these triggers must be designed so that indeed only ❵uninteresting✬ events get discarded. Finally, due in part to the nature of the underlying theory, the connection between theoretical predictions and recorded data is highly mediated (Morrison and Morgan, 1999), and analyses leading to discovery and measurement claims are intricate (Boge, 2021; Boge and Zeitnitz, 2021) .  \n* Forthcoming in: Duran, J. M. , and Pozzi, G. (Eds.), Philosophy of Science for Machine Learning, Synthese Library, Springer.  \n❸Institute for Philosophy and Political Science, TU Dortmund University, Emil-Figge-Str. 50, room 2 .247, 44227 Dortmund, Germany. ✢[orian-johannes.boge@udo.edu](orian-johannes.boge@udo.edu)  \n❹ Lamarr Institute for Machine Learning and Arti✜cial Intelligence  \n➓Institute for Science in Society, Chair for Philosophy of the Natural Sciences, Radboud University, Heyendaalseweg 135, 6525 AJ Nijmegen, [the Netherlands. henk.deregt@ru.nl](the Netherlands. henk.deregt@ru.nl)  \n1 See, e.g. , [https://home.cern/news/news/computing/cern-data-centre-passes-200-petabyte-milestone](https://home.cern/news/news/computing/cern-data-centre-passes-200-petabyte-milestone) (checked 05/23) .  \nGiven the sheer amounts of data and the complexities thus involved in data-generation, management and analysis, it is no wonder that ✏[p]article physicists began ✜ddling with arti✜cial intelligence (AI) in the late 1980s, just as the term ❵neural network✬ captured the public✬s imagination.✑ (Cho, 2017) Over these 40 or so years, there has been a steady co-evolution of particle physics (PP) and Machine Learning (ML) methods.2 Today, applications of ML in PP are vast and their number is ever-growing.3 What we here, in this chapter, will focus on are a few developments which suggest that genuine discoveries in PP seem possible with ML, and that these may outstrip physicists✬ present-day understanding of the sub-atomic domain. Thus, we will here confront the possibility of discovery without (human) understanding; a phenomenon which bears close connections to the pertinent topic of scienti✜c understanding✬s role in scienti✜c progress (Dellsén, 2021; de Regt, 2017; de Regt et al. , 2009; Grimm et al. , 2017; Rowbottom, 2023) and to the question of whether explanation falls behind in ML (Boge & Poznic, 2021; Boge et al. , 2022) .4  \nIn particular, we w","cbCaipqN1abCO5Uc","https://ap.wps.com/l/cbCaipqN1abCO5Uc","pdf",2233540,1,27,"English","en",105,"# Introduction\n## Data challenges in particle physics\n## ML applications in detectors, simulation, and analysis\n## Discovery of novel concepts and phenomena\n## Epistemic challenges and mitigation frameworks","[{\"question\":\"How is machine learning used in particle physics experiments today?\",\"answer\":\"Machine learning supports detector signal reconstruction, event simulation, and the determination of statistical ratios in final analyses.\"},{\"question\":\"What epistemic challenges arise when ML discovers concepts or phenomena?\",\"answer\":\"Discovery can outstrip scientists’ current understanding because understanding relies on theories, models, and concepts; lacking these concepts or theory, the resulting phenomena may be poorly understood.\"},{\"question\":\"What approaches are discussed to mitigate the impact of ML-based novel concept discoveries?\",\"answer\":\"The chapter discusses two recently proposed frameworks for learning from the machine, aimed at reducing the understanding advantage that ML may have over humans in these cases.\"}]","Machine Learning Discoveries and Scientific Understanding in Particle Physics - Problems and Prospects | PDF",1785939087,68,{"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},"machine-learning-discoveries-and-scientific-understanding-in-particle-physics-problems-and-prospects","",{"@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/machine-learning-discoveries-and-scientific-understanding-in-particle-physics-problems-and-prospects/127468/",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","2026-08-05",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},"How is machine learning used in particle physics experiments today?","Question",{"text":75,"@type":76},"Machine learning supports detector signal reconstruction, event simulation, and the determination of statistical ratios in final analyses.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What epistemic challenges arise when ML discovers concepts or phenomena?",{"text":80,"@type":76},"Discovery can outstrip scientists’ current understanding because understanding relies on theories, models, and concepts; lacking these concepts or theory, the resulting phenomena may be poorly understood.",{"name":82,"@type":73,"acceptedAnswer":83},"What approaches are discussed to mitigate the impact of ML-based novel concept discoveries?",{"text":84,"@type":76},"The chapter discusses two recently proposed frameworks for learning from the machine, aimed at reducing the understanding advantage that ML may have over humans in these cases.","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":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]