[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127653-en":3,"doc-seo-127653-105":31,"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},127653,962084925636,"Olivia Brown","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Machine Learning Interatomic Potentials to Predict Bond Dissociation Energies","Empirical force fields limit accuracy, transferability, and applicability to open-shell structures, while Machine Learning Interatomic Potentials (MLIPs) can act as accurate surrogates for ab initio potential energy surfaces. This dissertation develops transferable MLIPs usable for both closed- and open-shell molecules, extending scope to radical species for Bond Dissociation Energy (BDE) prediction. It compares GAP, ACE, and MACE for fitting, extrapolation to novel compounds, and BDE prediction, finding improved performance from MACE and limitations in joint GAP/ACE fitting. A transferable MACE model for C, H, O is evaluated on independent datasets and benchmarked against ANI-2x and AM1.","Machine Learning Interatomic Potentials to Predict Bond  \nDissociation Energies  \nElena Gelžinyt  \nSupervisor: Prof. Gábor Csányi  \nDepartment of Engineering  \nUniversity of Cambridge  \nThis dissertation is submitted for the degree of Doctor of Philosophy  \nSt Catharine’s College September 2023  \nDeclaration  \nThis thesis is the result of my own work and includes nothing that is the outcome of work done in collaboration except as specified in the text. The contents of this dissertation are not substantially the same as any work that has already been submitted before for any degree or other qualification. It does not exceed the prescribed word limit for the Engineering Degree Committee.  \nElena Gelžinyt September 2023  \nAcknowledgements  \nFirst, I would like to thank my supervisor, Professor Gábor Csányi, for the opportunity to join his research group and work on these exciting projects. His guidance, consideration and encouragement have helped me through the interesting and the challenging stretches and allowed me to learn so much meanwhile. I would also like to thank Dr Matthew D. Segall and Dr Mario Öeren for the origin of the Bond Dissociation Energy project and thoughtful feedback throughout. Special thanks to Mario, meetings with whom have always livened up my day. My sincere thanks are also to Dr Noam Berenstein for his help with and collaboration on the wfl package & article and for many patient pull request reviews. My similar thanks go to Professor Christoph Ortnerand Dr William (Chuck) Witt for their help with adjusting the interface to the ACE potential fitting code. I am also grateful to EPSRC and SynTech for their financial support and the opportunity to join the CDT programme.  \nAdditionally, I would like to express my appreciation for all of the support that has been available to me via the University Counselling Service and St Catharine’s College. Especially thanks to Anna, whose guidance has been instrumental in learning how to cope with the little and the big stresses of the final year.  \nI have been very fortunate to conduct my research alongside the interesting amazing fun bunch of people that the Csányi group & friends are. I thoroughly enjoyed our group meetings, excursions, sharing the office and wine & cheese. A special mention goes to Eszter and Will who have proofread parts of this thesis. My final warm thanks are to all my friends, Will and family. Thanks for being there and being great.  \nAbstract  \nEmpirical force fields are valuable tools in computational chemistry, however, they suffer from limitations in terms of accuracy, transferability and their lack of applicability to open-shell structures. Recently, Machine Learning Interatomic Potentials (MLIPs) have emerged as versatile surrogate models capable of accurately reproducing ab initio potential energy surfaces. However, most of their applications have been targeted at near-equilibrium closed-shell structures. This project aims to address this limitation by developing highly accurate and transferable MLIPs that can be applied to both closedand open-shell molecules. An accurate description of radical species extends the scope of possible applications to Bond Dissociation Energy (BDE) prediction, for example, with relevance to cytochrome P450 metabolism modelling. In this work, three methods are compared – Gaussian Approximation Potentials (GAP), Atomic Cluster Expansion (ACE), and MACE – in their ability to accurately fit closed-and open-shell hydrocarbon data, extrapolate to novel compounds and predict BDEs with required accuracy. The analysis reveals shortcomings in GAP and ACE when simultaneously fitting closed-and open-shell structures and demonstrates significantly better MACE performance when fitted to the same data. We further develop a transferable MACE model applicable to compounds containing carbon, hydrogen and oxygen chemical elements. To verify its transferability, we evaluate this model on several independent datasets and compare ","cbCaiaz7NVqo8QGE","https://ap.wps.com/l/cbCaiaz7NVqo8QGE","pdf",7612174,3,1,130,"English","en",105,"# Introduction\n## Outline of the Thesis\n# Methods\n## Regression\n## Fitting Potential Energy Surfaces\n## Reference Method\n## Calculating and Comparing Bond Dissociation Energies\n## Software\n# Challenges in Modelling Large Diverse Data Sets\n## The Hydrocarbon Dataset\n## Creating 3D Geometries\n## Iterative Data Collection and ACE Fittin","[{\"question\":\"What problem does the dissertation address about existing force fields?\",\"answer\":\"Empirical force fields can lack accuracy and transferability and often do not apply well to open-shell structures, limiting their use for systems involving radicals.\"},{\"question\":\"Which MLIP methods are compared for bond dissociation energy prediction?\",\"answer\":\"The work compares Gaussian Approximation Potentials (GAP), Atomic Cluster Expansion (ACE), and MACE, evaluating their ability to fit data, extrapolate, and predict BDEs with required accuracy.\"},{\"question\":\"What conclusions are drawn about model performance and transferability?\",\"answer\":\"Significant shortcomings are reported for GAP and ACE when fitting both closed- and open-shell structures, while MACE shows significantly better performance. A transferable MACE model for C/H/O is validated on independent datasets and benchmarks against ANI-2x and AM1, showing improved predicted BDE correlation.\"}]","Machine Learning Interatomic Potentials to Predict Bond Dissociation Energies | PDF",1785940521,328,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":29},"machine-learning-interatomic-potentials-to-predict-bond-dissociation-energies","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":20},"https://docshare.wps.com/document/research-report/",{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/machine-learning-interatomic-potentials-to-predict-bond-dissociation-energies/127653/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",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},"What problem does the dissertation address about existing force fields?","Question",{"text":76,"@type":77},"Empirical force fields can lack accuracy and transferability and often do not apply well to open-shell structures, limiting their use for systems involving radicals.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Which MLIP methods are compared for bond dissociation energy prediction?",{"text":81,"@type":77},"The work compares Gaussian Approximation Potentials (GAP), Atomic Cluster Expansion (ACE), and MACE, evaluating their ability to fit data, extrapolate, and predict BDEs with required accuracy.",{"name":83,"@type":74,"acceptedAnswer":84},"What conclusions are drawn about model performance and transferability?",{"text":85,"@type":77},"Significant shortcomings are reported for GAP and ACE when fitting both closed- and open-shell structures, while MACE shows significantly better performance. A transferable MACE model for C/H/O is validated on independent datasets and benchmarks against ANI-2x and AM1, showing improved predicted BDE correlation.","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":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]