[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-290881-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-290881-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","accelerating-lanthanide-and-actinide-opacity-calculations-with-machine-learning-models-student-thesis","Accelerating Lanthanide and Actinide Opacity Calculations with Machine-Learning Models - Student Thesis","","Lanthanides and actinides strongly shape the opacity of kilonova ejecta produced in neutron-star mergers, mainly through complex f-shell bound–bound line spectra that make atomic-structure opacity calculations computationally prohibitive. This thesis develops a two-stage machine-learning surrogate: a convolutional autoencoder compresses down-sampled opacity spectra into a latent space, then a gradient-boosted regressor reconstructs opacity. The model reproduces Planck-mean opacities with errors on the order of 10^3 cm2/g and cuts spectral RMSE by over 50% versus a nearest-neighbour baseline. A convolutional classifier additionally predicts element identity (99.7% accuracy), plasma temperature, and mass density, while evaluating opacities orders of magnitude faster. Performance declines under aggressive down-sampling and highly ionized regimes.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/accelerating-lanthanide-and-actinide-opacity-calculations-with-machine-learning-models-student-thesis/290881/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/accelerating-lanthanide-and-actinide-opacity-calculations-with-machine-learning-models-student-thesis/290881.png","ImageObject",300,407,{"name":42,"@type":43},"Skyler","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-20","2026-09-17",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",14,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why are lanthanide and actinide opacity calculations computationally challenging?","Question",{"text":63,"@type":64},"Their opacity is dominated by complex f-shell bound–bound line spectra, and conventional atomic-structure codes require extremely costly calculations.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How does the two-stage machine-learning approach work in this thesis?",{"text":68,"@type":64},"A convolutional autoencoder compresses down-sampled opacity spectra into a low-dimensional latent space, and a gradient-boosted regressor maps that latent representation back to opacity.",{"name":70,"@type":61,"acceptedAnswer":71},"What accuracy and speed improvements are reported compared with a nearest-neighbour baseline?",{"text":72,"@type":64},"The surrogate reproduces Planck-mean values with errors on the order of 10^3 cm2/g and reduces spectral RMSE by more than 50% versus the baseline, while evaluating opacities orders of magnitude faster.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},290881,1789642878,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":140,"language":141,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":12,"update_tm":80,"read_time":99,"preview_page_count":4},2336464648746,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Bachelor Thesis Physics and Astronomy  \nAccelerating Lanthanide and Actinide Opacity Calculations with Machine-Learning Models  \nBy:  \nKhoi Pham  \nConducted between March 31st, 2025, and June 19, 2025  \nStudent number: 14591103  \nResearch institute: ARCNL  \nSupervisor: Bowie Brewster & John Sheil  \nExaminer: Oscar Versolato  \nAbstract  \nLanthanides and actinides are heavy elements produced in neutron-star mergers that dominate the opacity of kilonova ejecta through complex f-shell line spectra, but calculating their bound–bound opacities with atomic-structure codes is computationally extremely costly. This work introduces a two-stage machine-learning approach that replaces those calculations. To address this, the NIST-LANL opacity database — comprising over 12000 high-resolution spectra spanning 28 elements, 27 temperatures and 17 densities — was collected into a unified dataset for machine-learning. A convolutional autoencoder first reduced a down-sampled 200-bin opacity spectrum to a low-dimensional latent space, which a gradient-boosted regressor then mapped back to opacity. This surrogate was able to reproduce Planck-mean values with errors on the order of 103 cm2 /g and reduced spectral RMSE by more than 50% relative to a nearest-neighbour baseline. Next, a convolutional classifier analyzed these spectra, inferring element identity with 99.7% accuracy, plasma temperature within 0.19 eV RMSE, and mass density within 8.35×105 RMSE. Although aggressive down-sampling smoothed out the narrowest resonances and performance degraded in the most highly ionized regimes, the surrogate evaluated opacities orders of magnitude faster than conventional atomic-structure codes. Future work on attention-based models to recover the narrow peaks, physics-informed loss functions to enforce atomic constraints, and uncertainty-driven active learning could sharpen accuracy and enable transparent, uncertainty-aware opacity calculations in nextgeneration astrophysical simulations.  \nContents  \n1 Introduction 1  \n1.1 Neutron Star Mergers and Kilonovae ..................... 1  \n1.2 Opacity in Heavy-Element Plasmas ...................... 1  \n1.3 The Planck Mean Opacity ........................... 1  \n1.4 The NIST-LANL Opacity database ...................... 2  \n1.5 Goals of this thesis ............................... 2  \n2 Machine Learning Models and Methodology 2  \n2.1 Introduction to Machine Learning ....................... 2  \n2.2 Database and pre-processing .......................... 3  \n2.2.1 Scaling experiments .......................... 3  \n2.2.2 Loss metrics ............................... 3  \n2.2.3 Binning ................................. 4  \n2.2.4 Training and test data ......................... 5  \n2.3 Regression models ............................... 5  \n2.3.1 Nearest-Neighbour Baseline ...................... 5  \n2.3.2 Fully Connected Neural Network (FC-NN) .............. 5  \n2.3.3 Convolutional Autoencoder (CAE) .................. 6  \n2.3.4 LightGBM Regression ......................... 7  \n2.3.5 Planck Mean Opacity Evaluation ................... 8  \n2.4 Classification models .............................. 8  \n2.4.1 Element Classification ......................... 8  \n2.4.2 Temperature and density classification ................ 9  \n3 Results 9  \n3.1 Baseline vs. Neural Nets vs. LightGBM (Regression) ............ 9  \n3.1.1 FC-NN Performance .......................... 9  \n3.1.2 Autoencoder Performance ....................... 10  \n3.1.3 LightGBM Performance ........................ 11  \n3.1.4 Overall comparison ........................... 12  \n3.2 Classification Results .............................. 13  \n4 Discussion 14  \n4.1 How Well Do ML Models Capture Spectral Features? ............ 14  \n4.2 Binning & Information Loss .......................... 15  \n4.3 Classification Takeaways ............................ 15  \n4.4 Future Extensions ............................... 16  \n4.4.1 Attention-Based Neural Networks ................... 16  \n4.4","cbCaiobSiQwlyTD8","https://ap.wps.com/l/cbCaiobSiQwlyTD8","pdf",1564634,24,"English","# Abstract\n# Contents\n## 1 Introduction\n## 2 Machine Learning Models and Methodology\n## 3 Results\n## 4 Discussion\n## 5 Conclusions","[{\"question\":\"Why are lanthanide and actinide opacity calculations computationally challenging?\",\"answer\":\"Their opacity is dominated by complex f-shell bound–bound line spectra, and conventional atomic-structure codes require extremely costly calculations.\"},{\"question\":\"How does the two-stage machine-learning approach work in this thesis?\",\"answer\":\"A convolutional autoencoder compresses down-sampled opacity spectra into a low-dimensional latent space, and a gradient-boosted regressor maps that latent representation back to opacity.\"},{\"question\":\"What accuracy and speed improvements are reported compared with a nearest-neighbour baseline?\",\"answer\":\"The surrogate reproduces Planck-mean values with errors on the order of 10^3 cm2/g and reduces spectral RMSE by more than 50% versus the baseline, while evaluating opacities orders of magnitude faster.\"}]","Accelerating Lanthanide and Actinide Opacity Calculations with Machine-Learning Models - Student Thesis | PDF"]