[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121354-en":3,"doc-seo-121354-105":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},121354,687197207057,"Sage","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Prediction of Exotic (n,2n) Cross Sections using a Regression Tree Machine Learning Algorithm","XGBoost machine learning regression is employed to predict (n,2n) microscopic cross sections by training on physical parameters for target nuclei and their evaluated (n,2n) cross sections from ENDF/B-VIII. Models are developed for nuclides with 30 ≤ A ≤ 208 and predictions are benchmarked against multiple evaluated nuclear data libraries including ENDF/B-VIII, JENDL-5, JEFF-3.3, CENDL-3.2, and TENDL-2021. For 73.5 ± 1.0% of nuclides in ENDF/B-VIII, agreement is achieved with r2 ≥ 0.95. The method is then extended to a broad set of exotic nuclides and compared with TENDL-2021 and JENDL-5 evaluations.","Prediction of Exotic (n,2n) Cross Sections using a Regression Tree Machine  \nLearning Algorithm  \nRohan Teelock-Gayaa , Valeria Raffuzzia , Eugene Shwagerausa , Lee Morganb  \na Department of Engineering, University of Cambridge, Trumpington Street, Cambridge, CB2 1PZ, United Kingdom bAWE Aldermaston, Reading, RG7 4PR, United Kingdom  \nAbstract  \nThe XGBoost machine learning algorithm for regression, was used to predict (n,2n) microscopic cross sections, by training models on physical parameters describing various target nuclei, and their corresponding evaluated (n,2n) cross sections sourced from ENDF/B-VIII. Research was concentrated on nuclides with nucleon numbers 30 ≤ A ≤ 208. Machine learning predictions were compared to library evaluations from ENDF/B-VIII, JENDL-5, JEFF-3.3, CENDL-3.2, and TENDL-2021 . Predictions for many nuclides were found to be in agreement with existing evaluated cross sections, with an r2 ≥ 0.95 with respect to at least one library evaluation found for 73 .5 ± 1.0% of nuclides in ENDF/B-VIII. Predictions were subsequently made on a wide range of exotic nuclides, and compared to evaluations from the TENDL-2021 and JENDL-5 libraries.  \nKeywords:  \nnuclear data, machine learning,(n,2n) reaction  \nI. Introduction  \nNuclear reaction cross sections, σ(E), are an important subset of nuclear data used in scientific and engineering applications. Uses for such data include designing and modelling nuclear systems, where cross sections represent the likelihood of neutrons undergoing various reactions. One such reaction is the (n,2n) reaction channel, a fast, threshold reaction, with its cross sections exhibiting no resonances, and a single peak. (n,2n) reactions are especially relevant to fusion systems and for general radiation transport (1) in any primarily fast neutron system, e.g. fusion reactor modelling, medical isotope production, astrophysics, and radiation shielding calculations. Conventionally, cross section data used for these applications are generated by nuclear scientists, also known as nuclear data evaluators, through the nuclear data evaluation process. Evaluators consider both experimental measurements of cross sections typically sourced from the EXFOR library, and theoretical predictions of cross section behaviour calculated using physics models of nuclear reactions, such as Hauser-Feshbach theory (2) and the optical model (3) .  \nAn important blind spot of experimental methods is the production of cross sections for exotic nuclides. Exotic nuclides are unstable nuclides with typically large neutron/proton asymmetry and short half-lives (4) with their cross section data being of interest to astrophysics research and fast neutron systems modelling (5; 6) .. Their instability, and often the difficulty of producing exotic nuclides, makes it challenging to measure exotic cross sections experimentally. Few facilities, such as the FRIB in Michigan, are capable of measuring exotic cross sections (7) . Additionally, the large number of unstudied exotic nuclides serves as a bottleneck for exotic cross section production. There are over a thousand exotic nuclides whose cross sections have not been measured experimentally, and a multitude of reaction channels possible for each of these nuclides, e.g. (n,2n) and (n,γ) . For nuclides that are observed experimentally, measurements often have large uncertainties, or contradict other experimental campaigns or theoretical models of reactions. Hence, the weighting of such measurements must be decided by evaluators. This is one potential mechanism for the introduction of ”evaluator bias”, during the production of evaluated cross sections. Hence theoretical models, such as the ones employed by the physics codes TALYS (8) and EMPIRE (9), which make use of the optical model and Hauser-Feshbach theory, are generally the only available method of producing cross sections for exotic nuclei. Consequently, the TENDL-2021 library (10) produced using TALYS serves as the m","cbCaijDFPwaHFjXj","https://ap.wps.com/l/cbCaijDFPwaHFjXj","pdf",8254883,1,21,"English","en",105,"# Abstract\n# Keywords\n# I. Introduction\n## Nuclear reaction cross sections and (n,2n) relevance\n## Challenges for exotic nuclides and evaluated libraries\n## Motivation for machine learning in nuclear data","[{\"question\":\"Which machine learning algorithm is used to predict (n,2n) microscopic cross sections?\",\"answer\":\"The study uses the XGBoost machine learning algorithm for regression, implemented via a regression-tree approach.\"},{\"question\":\"What data sources are used for training and benchmarking the (n,2n) cross-section predictions?\",\"answer\":\"Training uses physical parameters linked with evaluated (n,2n) cross sections from ENDF/B-VIII, and benchmarking compares against ENDF/B-VIII, JENDL-5, JEFF-3.3, CENDL-3.2, and TENDL-2021.\"},{\"question\":\"How does the work validate predictive performance and what proportion of nuclides shows strong agreement?\",\"answer\":\"Agreement is assessed using r2, with r2 ≥ 0.95 achieved for 73.5 ± 1.0% of nuclides when compared to at least one library evaluation from ENDF/B-VIII.\"}]","Prediction of Exotic (n,2n) Cross Sections using a Regression Tree Machine Learning Algorithm | PDF",1785735210,53,{"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},"prediction-of-exotic-n2n-cross-sections-using-a-regression-tree-machine-learning-algorithm","",{"@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/prediction-of-exotic-n2n-cross-sections-using-a-regression-tree-machine-learning-algorithm/121354/",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","2026-08-03",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},"Which machine learning algorithm is used to predict (n,2n) microscopic cross sections?","Question",{"text":75,"@type":76},"The study uses the XGBoost machine learning algorithm for regression, implemented via a regression-tree approach.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What data sources are used for training and benchmarking the (n,2n) cross-section predictions?",{"text":80,"@type":76},"Training uses physical parameters linked with evaluated (n,2n) cross sections from ENDF/B-VIII, and benchmarking compares against ENDF/B-VIII, JENDL-5, JEFF-3.3, CENDL-3.2, and TENDL-2021.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the work validate predictive performance and what proportion of nuclides shows strong agreement?",{"text":84,"@type":76},"Agreement is assessed using r2, with r2 ≥ 0.95 achieved for 73.5 ± 1.0% of nuclides when compared to at least one library evaluation from ENDF/B-VIII.","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"]