[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-290455-105":59,"doc-detail-290455-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","the-landscape-of-unfolding-with-machine-learning","The landscape of unfolding with machine learning","","Recent innovations from machine learning enable data unfolding without binning while incorporating correlations across many dimensions. A comprehensive set of established, upgraded, and newly proposed ML-based unfolding methods is presented and benchmarked using the same two datasets. The study shows that all evaluated techniques can accurately reproduce particle-level spectra over complex observables. Owing to their conceptual diversity, the approaches provide a versatile toolkit for precision measurements probing the Standard Model with enhanced sensitivity.",{"@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/the-landscape-of-unfolding-with-machine-learning/290455/",{"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/the-landscape-of-unfolding-with-machine-learning/290455.png","ImageObject",300,407,{"name":92,"@type":93},"Hazel","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-07","2026-09-17",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",12,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does ML-based unfolding address compared with forward inference?","Question",{"text":113,"@type":114},"Unfolding estimates pre-detector particle-level distributions directly from data, avoiding the need for detector simulation and enabling analysis by a broader community beyond a specific experiment’s setup.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"What distinguishes the ML unfolding methods discussed in this paper?",{"text":118,"@type":114},"The paper benchmarks a diverse set of ML approaches, including reweighting methods like OmniFold, distribution mapping methods such as Schrödinger Bridge and Direct Diffusion, and generative unfolding with several architectures.",{"name":120,"@type":111,"acceptedAnswer":121},"How is the performance of the methods evaluated?",{"text":122,"@type":114},"All approaches are evaluated on the same two datasets, and they are shown to reproduce particle-level spectra across complex observables accurately.","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},290455,1790441636,{"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":145,"read_time":146,"preview_page_count":4},137441390410,"https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984","UC Irvine  \nUC Irvine Previously Published Works  \nTitle  \nThe landscape of unfolding with machine learning  \nPermalink  \n[https://escholarship.org/uc/item/7gg776x9](https://escholarship.org/uc/item/7gg776x9)  \nJournal  \nSciPost Physics, 18(2)  \nISSN  \n2542-4653  \nAuthors  \nHuetsch, Nathan  \nMariño Villadamigo, Javier Shmakov, Alexander et al.  \nPublication Date  \n2025  \nDOI  \n10.21468/scipostphys.18.2.070  \nCopyright Information  \nThis work is made available under the terms of a Creative Commons Attribution License, available at [https://creativecommons.org/licenses/by/4.0/](https://creativecommons.org/licenses/by/4.0/)  \nPeer reviewed  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \narXiv :2404 . 18807v2 [hep-ph] 17 May 2024  \nThe Landscape of Unfolding with Machine Learning  \nNathan Huetsch1 , Javier Mariño Villadamigo1 , Alexander Shmakov2 , Sascha Diefenbacher3 , Vinicius Mikuni3 , Theo Heimel1 , Michael Fenton2 , Kevin Greif2 , Benjamin Nachman3,4 , Daniel Whiteson2 , Anja Butter1,5 , and Tilman Plehn1,6  \n1 Institut für Theoretische Physik, Universität Heidelberg, Germany  \n2 Department of Physics and Astronomy, University of California, Irvine, USA  \n3 Physics Division, Lawrence Berkeley National Laboratory, Berkeley, USA  \n4 Berkeley Institute for Data Science, University of California, Berkeley, USA  \n5 LPNHE, Sorbonne Université, Université Paris Cité, CNRS/IN2P3, Paris, France  \n6 Interdisciplinary Center for Scientific Computing (IWR), Universität Heidelberg, Germany  \nMay 20, 2024  \nAbstract  \nRecent innovations from machine learning allow for data unfolding, without binning and including correlations across many dimensions. We describe a set of known, upgraded, and new methods for ML-based unfolding. The performance of these approaches are evaluated on the same two datasets. We find that all techniques are capable of accurately reproducing the particle-level spectra across complex observables. Given that these approaches are conceptually diverse, they offer an exciting toolkit for a new class of measurements that can probe the Standard Model with an unprecedented level of detail and may enable sensitivity to new phenomena.  \n\n| Contents\u003Cbr>1 Introduction\u003Cbr>2 ML-Unfolding\u003Cbr>2.1 Reweighting: (b)OmniFold\u003Cbr>2.2 Mapping distributions: Schrödinger Bridge and Direct Diffusion\u003Cbr>2.3 Generative unfolding: cINN, Transfermer, CFM, TraCFM, Latent Diffusion\u003Cbr>3 Detector unfolding: Z+jets\u003Cbr>3.1 Data and preprocessing\u003Cbr>3.2 Reweighting\u003Cbr>3.3 Mapping distributions\u003Cbr>3.4 Generative unfolding\u003Cbr>3.5 Learned event migration\u003Cbr>4 Unfolding to parton level: top pairs\u003Cbr>4.1 Data\u003Cbr>4.2 Generative unfolding\u003Cbr>4.3 Generative unfolding using physics\u003Cbr>5 Outlook\u003Cbr>A Combined Z+jets results\u003Cbr>B Hyperparameters\u003Cbr>References | 3\u003Cbr>4 4\u003Cbr>5 8\u003Cbr>14\u003Cbr>14\u003Cbr>15\u003Cbr>17\u003Cbr>19\u003Cbr>19\u003Cbr>22\u003Cbr>22\u003Cbr>22\u003Cbr>24\u003Cbr>26\u003Cbr>28\u003Cbr>29\u003Cbr>30 |\n| --- | --- |\n\n1 Introduction  \nParticle physics experiments seek to reveal clues about the fundamental properties of particles and their interactions. A key challenge is that predictions from quantum field theory are atthe level of partons, while experiments observe the corresponding detector signatures. Precise and detailed simulations link these two levels [1] . They fold predictions for the hard process through QCD effects, hadronization, and the detector response to compare with data. This statistically powerful forward inferences approach has been widely used.  \nHowever, forward inference requires access to the data and the detector simulation. These conditions are rarely satisfied outside of a given experiment, severely limiting the ability of the broader community to study particle physics data. In addition, analysis of data from the highluminosity LHC with forward inference will require precise simulations for every hypothesis, challenging available computing resources.  \nAn alternative approach is unfolding. Rather than correcting pr","cbCaioJGUIQET0CG","https://ap.wps.com/l/cbCaioJGUIQET0CG","pdf",1773083,36,"English","# 1 Introduction\n# 2 ML-Unfolding\n## 2.1 Reweighting: (b)OmniFold\n## 2.2 Mapping distributions: Schrödinger Bridge and Direct Diffusion\n## 2.3 Generative unfolding: cINN, Transfermer, CFM, TraCFM, Latent Diffusion\n# 3 Detector unfolding: Z+jets\n## 3.1 Data and preprocessing\n## 3.2 Reweighting\n## 3.3 Mapping distributions\n## 3.4 Generative unfolding\n## 3.5 Learned event migration\n# 4 Unfolding to parton level: top pairs\n## 4.1 Data\n## 4.2 Generative unfolding\n## 4.3 Generative unfolding using physics\n# 5 Outlook\n# A Combined Z+jets results\n# B Hyperparameters\n# References","[{\"question\":\"What problem does ML-based unfolding address compared with forward inference?\",\"answer\":\"Unfolding estimates pre-detector particle-level distributions directly from data, avoiding the need for detector simulation and enabling analysis by a broader community beyond a specific experiment’s setup.\"},{\"question\":\"What distinguishes the ML unfolding methods discussed in this paper?\",\"answer\":\"The paper benchmarks a diverse set of ML approaches, including reweighting methods like OmniFold, distribution mapping methods such as Schrödinger Bridge and Direct Diffusion, and generative unfolding with several architectures.\"},{\"question\":\"How is the performance of the methods evaluated?\",\"answer\":\"All approaches are evaluated on the same two datasets, and they are shown to reproduce particle-level spectra across complex observables accurately.\"}]","The landscape of unfolding with machine learning | PDF",1789641728,91]