[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-118035-en":3,"doc-seo-118035-105":30,"detail-sidebar-cat-0-en-105":92},{"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":20,"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},118035,4398048949847,"Eliana","https://ap-avatar.wpscdn.com/avatar/400002536579ef2da7f?_k=1778318612642679267",8,"Research & Report","Machine Learning Force Fields for Molecular Chemistry - Doctoral Dissertation - January 2024","Machine learning driven force fields are developed to overcome the limits of electronic-structure simulations that are typically accurate only for small systems and short times. Traditional force fields enable fast modelling by bypassing explicit electron treatment, while ML approaches aim to reproduce electronic structure results with high accuracy. The thesis introduces ACE basis functions for accurate custom force fields of small molecules, extends them for many chemical elements, then proposes multi-ACE and MACE as a unifying, robust framework, and finally presents MACE-OFF23 for transferable organic molecular modelling in vacuum and condensed phases.","Machine Learning Force Fields for Molecular Chemistry  \nDávid Péter Kovács  \nDepartment of Engineering  \nUniversity of Cambridge  \nThis dissertation is submitted for the degree of Doctor of Philosophy  \nSt Catharine’s College January 2024  \nDeclaration  \nI hereby declare that except where specific reference is made to the work of others, the contents of this dissertation are original and have not been submitted in whole or in part for consideration for any other degree or qualification in this, or any other university. This dissertation is my own work and contains nothing which is the outcome of work done in collaboration with others, except as specified in the text and Acknowledgements. This dissertation contains fewer than 65,000 words including appendices, bibliography, footnotes, tables and equations and has fewer than 150 figures.  \nDávid Péter Kovács  \nJanuary 2024  \nAbstract  \nThe first principles computational modelling of molecular systems is a long-standing pursuit in the scientific community. It has traditionally been tackled by developing approximate solutions to quantum mechanics. Simulations using these electronic structure based methods can be highly accurate but are limited to small system sizes or short time scales. The traditional alternative is force fields that enable fast and accurate simulations by bypassing the treatment of the electrons and describing the system solely in terms of the atomic positions. The emergence of machine learning tools has opened up the opportunity for the development of high accuracy force fields trained directly to reproduce the results of electronic structure calculations.  \nThis thesis presents new developments that lead to improved machine learning force fields for molecular chemistry. Firstly, a set of linearly complete basis functions, called ACE, is demonstrated to yield high accuracy custom made force fields for small molecules. By recognising the symmetric tensor structure of these basis functions, the framework is extended to enable the simultaneous description of a large number of chemical elements.  \nNext, multi-ACE is proposed, which provides a unifying theory of most classical and machine learning force fields. Using the design space set out in this theory, a new method, called MACE is created. MACE is shown to provide simple, robust, accurate, and efficient force fields for a wide range of molecular systems. Finally, MACE-OFF23 a new transferable force field for organic molecules is proposed and demonstrated to be capable of accurately describing not only molecules in vacuum but also in the condensed phase.  \nAcknowledgements  \nFirst, I would like to express my gratitude to my supervisor Prof. Gábor Csányi for his encouragement, advice, and guidance throughout my PhD. In particular, I would like to thank him for allowing me to pursue my interests freely and to allow me to choose topics that were of interest to me.  \nI would also like to thank the members of the Gabor group for providing an intellectually stimulating and fun environment to work in. In particular, I would like to thank Ilyes Batatia, with whom I was fortunate to work on many of my projects. I also thank Cas van der Oord for introducing me to ACE in the initial days of my PhD. I am also grateful for many fruitful discussions around the coffee machine with William Baldwin and James Darby. I would also like to thank Harry Moore for helping me with biomolecular simulations.  \nFurthermore, I would like to express my gratitude to the many external collaborators I had, including Chrsitoph Ortner for the guidance in the mathematical parts of my PhD, Daniel Cole for discussing with me throughout my PhD and inspiring me to work on difficult problems relevant to drug discovery, Joshua Horton for helping with the development and testing of new organic force fields, Venkat Kapil for introducing me to quantum dynamicsand using ML force fields for spectroscopy, Gus Hart for providing valuable data and ideas for testing the TrACE ar","cbCaitAEgN7hOfij","https://ap.wps.com/l/cbCaitAEgN7hOfij","pdf",15310046,1,142,"English","en",105,"# Introduction\n## Outline and Key References\n# Background\n## Electronic Structure Methods-The Ground Truth\n## The Potential Energy Surface\n## Atomistic Modelling Using Force Fields\n## Overview of Force Field Methods for Molecular Chemistry\n## Benchmark Datasets for Comparing Molecular Force Fields","[{\"question\":\"Why are machine learning force fields needed in molecular chemistry?\",\"answer\":\"Electronic-structure simulations are accurate but constrained by system size and timescale. Force fields enable faster simulations, and machine learning aims to reach high accuracy by learning from electronic-structure results.\"},{\"question\":\"What does ACE contribute to improved force field accuracy?\",\"answer\":\"ACE uses linearly complete basis functions to build high-accuracy, custom-made force fields for small molecules. The framework is extended by exploiting the symmetric tensor structure to cover many chemical elements.\"},{\"question\":\"How do MACE and MACE-OFF23 extend transferable modelling?\",\"answer\":\"Multi-ACE provides a unifying theory that guides a new method called MACE, which delivers simple, robust, accurate, and efficient force fields across molecular systems. MACE-OFF23 is a transferable organic force field shown to work both in vacuum and in the condensed phase.\"}]","Machine Learning Force Fields for Molecular Chemistry - Doctoral Dissertation - January 2024 | PDF",1785680907,358,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-force-fields-for-molecular-chemistry-doctoral-dissertation-january-2024","",{"@graph":36,"@context":86},[37,54,69],{"@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/machine-learning-force-fields-for-molecular-chemistry-doctoral-dissertation-january-2024/118035/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-02",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"Why are machine learning force fields needed in molecular chemistry?","Question",{"text":76,"@type":77},"Electronic-structure simulations are accurate but constrained by system size and timescale. Force fields enable faster simulations, and machine learning aims to reach high accuracy by learning from electronic-structure results.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What does ACE contribute to improved force field accuracy?",{"text":81,"@type":77},"ACE uses linearly complete basis functions to build high-accuracy, custom-made force fields for small molecules. The framework is extended by exploiting the symmetric tensor structure to cover many chemical elements.",{"name":83,"@type":74,"acceptedAnswer":84},"How do MACE and MACE-OFF23 extend transferable modelling?",{"text":85,"@type":77},"Multi-ACE provides a unifying theory that guides a new method called MACE, which delivers simple, robust, accurate, and efficient force fields across molecular systems. MACE-OFF23 is a transferable organic force field shown to work both in vacuum and in the condensed phase.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]