[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-290399-105":59,"doc-detail-290399-en":130},{"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":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","multi-fidelity-machine-learning-for-excited-state-energies-of-molecules","Multi-Fidelity Machine Learning for Excited State Energies of Molecules","","Accurate yet fast prediction of molecular excited-state energies remains highly challenging, especially when detailed energy funnels in larger molecular aggregates are required. Machine learning can help, but the cost of generating high-accuracy training datasets is a major obstacle. This work introduces multi-fidelity machine learning that combines a small set of highly accurate, expensive data with larger amounts of cheaper, lower-accuracy data to achieve high accuracy at lower cost. The method predicts first excited-state energies for benzene, naphthalene, and anthracene across conformations from classical molecular dynamics and real-time DFTB. Benchmark results show over a 30× reduction in data-generation effort while matching single-fidelity high-cost accuracy.",{"@graph":69,"@context":122},[70,84,105],{"@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/multi-fidelity-machine-learning-for-excited-state-energies-of-molecules/290399/",{"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/multi-fidelity-machine-learning-for-excited-state-energies-of-molecules/290399.png","ImageObject",300,407,{"name":92,"@type":93},"Eliana","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-17","2026-09-17",true,{"@type":102,"interactionType":103,"userInteractionCount":34},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does the paper address in predicting molecular excited states?","Question",{"text":112,"@type":113},"It addresses the difficulty of obtaining accurate excited-state energies with fast computation, where the main bottleneck is the high cost of producing high-accuracy training data.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does multi-fidelity machine learning reduce training cost?",{"text":117,"@type":113},"It combines a small number of highly accurate (costly) training samples with a larger set of less accurate (cheaper) samples to reach the accuracy associated with high-cost data while lowering overall computational effort.",{"name":119,"@type":110,"acceptedAnswer":120},"Which molecules and computational sources are used to evaluate the approach?",{"text":121,"@type":113},"The approach is tested on benzene, naphthalene, and anthracene, using conformations from classical molecular dynamics simulations and excited-state evaluations from real-time density functional tight-binding calculations.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},290399,1790527352,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":34,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145,"preview_page_count":4},4398048949847,"https://ap-avatar.wpscdn.com/avatar/400002536579ef2da7f?_k=1778318612642679267","arXiv :2305 . 11292v1 [physics .chem-ph] 18 May 2023  \nMulti-Fidelity Machine Learning for Excited State Energies of Molecules  \nVivin Vinod 1 , Sayan Maity2 , Peter Zaspel 1* and Ulrich Kleinekath􀁿ofer2*  \n1 School of Computer Science and Engineering, Constructor University, Campus Ring 1, 28759 Bremen, Germany.  \n2 School of Science, Constructor University, Campus Ring 1, 28759 Bremen, Germany.  \n*Corresponding author(s) . E-mail(s): [pzaspel@constructor.university](pzaspel@constructor.university) ; [ukleinekathoefer@constructor.university](ukleinekathoefer@constructor.university) ;  \nAbstract  \nThe accurate but fast calculation of molecular excited states is still a very challenging topic. For many applications, detailed knowledge of the energy funnel in larger molecular aggregates is of key importance requiring highly accurate excited state energies. To this end, machine learning techniques can be an extremely useful tool thoughthe cost of generating highly accurate training datasets still remains a severe challenge. To overcome this hurdle, this work proposes the use of multi-􀀌delity machine learning where very little training data from high accuracies is combined with cheaper and less accurate data to achieve the accuracy of the costlier level. In the present study, the approach is employed to predict the 􀀌rst excited state energies for three molecules of increasing size, namely, benzene, naphthalene, and anthracene. The energies are trained and tested for conformations stemming from classical molecular dynamics simulations and from real-time density functional tight-binding calculations. It can be shown that the multi-􀀌delity machine learning model can achieve the same accuracy asa machine learning model built only on high cost training data while having a much lower computational e􀀋ort to generate the data. The numerical gain observed in these benchmark test calculations was over a factor of 30 but certainly can be much higher for high accuracy data.  \n2 MFML for Excited States  \nFig. 1: Multi-􀀌delity machine learning (MFML) based on kernel ridge regression signi􀀌cantly reduces the cost of training a machine learning model for the prediction of quantum chemistry properties, here, excited state energies. In contrast to the conventional single-􀀌delity machine learning method, the discussed method uses data from multiple 􀀌delities with a few highly accurate (and costly) data samples and a growing number of less accurate (hence usually cheaper) data samples, thereby reducing the overall computational cost for the generation of the training data. This procedure helps to expedite the machine learning pipeline for predicting (􀀌rst) excited state energies.  \nKeywords: excited state energies, machine learning, multi-􀀌delity machine learning, kernel ridge regression, electronic structure theory  \n1 Introduction  \nExcited states form the basis for understanding various photo-induced processes in physics, chemistry, and the life sciences. A detailed knowledge of their energies and properties is key in uncovering the secrets of the intricate working of many systems. Moreover, in many technical applications such as photovoltaics or light-emitting diodes, excited states play a key role as well. As one particular example, we would like to mention the collection of solar energy by light-harvesting complexes not only in plants and algae but also in  \nMFML for Excited States 3  \nsome bacteria [1] . To be more speci􀀌c, a recent model of a light-harvesting organelle, a chromatophore, includes more than 2000 pigment molecules. In order to determine the 􀀍ow of excitation energy in such a system or to study its spectroscopic properties, one usually needs to determine the time evolution of the excited states for this large number of molecules [2{4] . At the same time, however, the determination of the excited state energies needs to be rather accurate since small di􀀋erences in energy can in􀀍uence the direction of energy 􀀍ow which might be cruc","cbCaiff9Q2itNJnm","https://ap.wps.com/l/cbCaiff9Q2itNJnm","pdf",769355,26,"English","# Abstract\n# Introduction\n## Excited states and computational cost\n## Machine learning for excited-state calculations","[{\"question\":\"What problem does the paper address in predicting molecular excited states?\",\"answer\":\"It addresses the difficulty of obtaining accurate excited-state energies with fast computation, where the main bottleneck is the high cost of producing high-accuracy training data.\"},{\"question\":\"How does multi-fidelity machine learning reduce training cost?\",\"answer\":\"It combines a small number of highly accurate (costly) training samples with a larger set of less accurate (cheaper) samples to reach the accuracy associated with high-cost data while lowering overall computational effort.\"},{\"question\":\"Which molecules and computational sources are used to evaluate the approach?\",\"answer\":\"The approach is tested on benzene, naphthalene, and anthracene, using conformations from classical molecular dynamics simulations and excited-state evaluations from real-time density functional tight-binding calculations.\"}]","Multi-Fidelity Machine Learning for Excited State Energies of Molecules | PDF",1789641614,66]