[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-466256-105":59,"doc-detail-466256-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","methods-in-pes-learn-direct-fit-machine-learning-of-bornoppenheimer-potential-energy-surfaces","Methods in PES-Learn - Direct-Fit Machine Learning of Born–Oppenheimer Potential Energy Surfaces","","PES-LEARN version 1.0 is presented as an open-source software package for automatically constructing machine learning models for semi-global molecular potential energy surfaces (PESs). Interoperability is enhanced through a new Python API that streamlines workflows via QCSchema-based input and output. The article introduces kernel ridge regression (KRR) within PES-LEARN, evaluates selected semi-global PES capabilities, and benchmarks all available ML methods using benzene and ethanol datasets from rMD17, including fitting performance and timing. Gradient prediction via neural network models for ethanol and benzene is also demonstrated and benchmarked. The project remains active and invites community contributions.",{"@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/methods-in-pes-learn-direct-fit-machine-learning-of-bornoppenheimer-potential-energy-surfaces/466256/",{"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/methods-in-pes-learn-direct-fit-machine-learning-of-bornoppenheimer-potential-energy-surfaces/466256.png","ImageObject",300,407,{"name":92,"@type":93},"Cart","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-16","2026-09-30",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 is PES-LEARN and what problem does it target?","Question",{"text":112,"@type":113},"PES-LEARN is an open-source package that automates dataset generation and machine learning model construction for direct fitting of Born–Oppenheimer potential energy surfaces.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What new capability does this paper add to PES-LEARN?",{"text":117,"@type":113},"It adds kernel ridge regression (KRR) as a new machine learning method and improves interoperability with a Python API integrated with QCSchema workflows.",{"name":119,"@type":110,"acceptedAnswer":120},"How are the machine learning methods benchmarked in the paper?",{"text":121,"@type":113},"The paper benchmarks fitting performance and timings using benzene and ethanol datasets from the rMD17 database, and it also evaluates gradient prediction with neural network models for these systems.","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},466256,1791696136,{"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":56,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":143,"read_time":144,"preview_page_count":19,"preview_urls":145},18829141979164,"https://eur-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8","Article  \nMethods in PES-Learn: Direct-Fit Machine Learning of Born–Oppenheimer Potential Energy Surfaces  \nIan T. Beck , Justin M. Turney and Henry F. Schaefer III *  \nAcademic Editor: Maxim L. Kuznetsov  \nReceived: 3 December 2025  \nRevised: 19 December 2025  \nAccepted: 19 December 2025  \nPublished: 25 December 2025  \nCopyright: © 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.  \nDepartment of Chemistry, Center for Computational Quantum Chemistry, University of Georgia, Athens, GA 30602, USA  \n* Correspondence: [ccq@uga.edu](ccq@uga.edu)  \nAbstract  \nThe release of PES-LEARN version 1.0 as an open-source software package for the automatic construction of machine learning models of semi-global molecular potential energy surfaces (PESs) is presented. Improvements to PES-LEARN’s interoperability are stressed with new Python API that simplifies workflows for PES construction via interaction with QCSchema input and output infrastructure. In addition, a new machine learning method is introduced to PES-LEARN: kernel ridge regression (KRR) . The capabilities of KRR are emphasized with examination of select semi-global PESs. All machine learning methods available in PESLEARN are benchmarked with benzene and ethanol datasets from the rMD17 database to illustrate PES-LEARN’s performance ability. Fitting performance and timings are assessed for both systems. Finally, the ability to predict gradients with neural network models is presented and benchmarked with ethanol and benzene. PES-LEARN is an active project and welcomes community suggestions and contributions.  \nKeywords: machine learning; potential energy surface; kernel ridge regression; neural networks  \n1. Introduction  \nThe concept of a Born–Oppenheimer potential energy surface (PES) is fundamental to a multitude of chemical disciplines. An abundance of chemical information arises from the PES, including but not limited to reaction dynamics, kinetics, and geometric stationary-point information [1–5] . When attempting to model a PES, construction is often limited by the quantity of stationary-point information that is required. To alleviate the burden of informational requirements in PES construction, recent developments have turned toward fitting mathematical expressions to describe PESs in order to reduce the cost of construction [6] . In particular, machine learning (ML) methods have become popular for reducing the cost of modeling a PES [7–16] . A number of packages have been developed to facilitate the construction of ML models to accurately describe molecular PES and related chemical properties while reducing overall computational cost [17–24] . Several such packages have been highlighted in a recent special issue in The Journal of Chemical Physics for “Software for Atomistic Machine Learning” such as PANNA, MLIP-3, and DeePMD-kit, among others [25–27] .  \nMachine learning methods that describe PESs are often broadly categorized into two groups. The first group attempts to describe the total energy of a system as the sum of atomistic energy contributions. These methods benefit in that they are able to describe an almost arbitrary sized system; however, these methods typically require a large amount  \nof training data [17,28–32] . The second group of methods, which are the focus of this paper, aim to fit system-specific PESs directly. While atomistic methods are able to fit system specific PESs as well, the advantage of direct-fit methods is the fewer training points required and faster training time compared to atomistic methods [33] . The limitations, however, include the size of the system that is able to be studied (typically ≤ 10 atoms) and non-transferability to other systems.  \nDirect-fit ML has been repeatedly demonstrated as an accurate and reliable method for the construction of system-specific PESs. Popular ML methods that ","cbCaicdYsmRQe1zt","https://ap.wps.com/l/cbCaicdYsmRQe1zt","pdf",894620,"English","# Abstract\n# Introduction\n## Born–Oppenheimer PES and modeling motivation\n## Two categories of ML PES modeling\n## Direct-fit methods and prior benchmarks\n## PES-LEARN software capabilities and scope","[{\"question\":\"What is PES-LEARN and what problem does it target?\",\"answer\":\"PES-LEARN is an open-source package that automates dataset generation and machine learning model construction for direct fitting of Born–Oppenheimer potential energy surfaces.\"},{\"question\":\"What new capability does this paper add to PES-LEARN?\",\"answer\":\"It adds kernel ridge regression (KRR) as a new machine learning method and improves interoperability with a Python API integrated with QCSchema workflows.\"},{\"question\":\"How are the machine learning methods benchmarked in the paper?\",\"answer\":\"The paper benchmarks fitting performance and timings using benzene and ethanol datasets from the rMD17 database, and it also evaluates gradient prediction with neural network models for these systems.\"}]","Methods in PES-Learn - Direct-Fit Machine Learning of Born–Oppenheimer Potential Energy Surfaces | PDF",1790770880,48,[146,150,154,158],{"page_no":8,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/182bfc78afbd176d8685770726700094.webp",983,1390,{"page_no":14,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/0798012cbd48aeb3e2400ee1eebe0d04.webp",982,1388,{"page_no":81,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/896b5b7b0147bc1849785ac518de120e.webp",981,1387,{"page_no":19,"url":159,"width":160,"height":161},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/83bfc484cd91371eba2d4018fff3354c.webp",980,1385]