[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122344-en":3,"doc-seo-122344-105":30,"detail-sidebar-cat-0-en-105":91},{"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},122344,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Expanding the frontiers of glass science: new insights derived from machine learning approaches - Doctor of Philosophy thesis","Glasses, as amorphous solids, offer versatile performance across many applications and benefit from compositional flexibility beyond crystalline stoichiometric constraints. Although this creates a vast design space, only a limited set of glass compositions has been synthesized historically, leaving major opportunities for materials with improved properties and new functions. The work integrates molecular dynamics simulations and machine learning to efficiently map composition–property relationships, accelerating novel glass discovery through analytical topological modeling, mechanical property extrapolation, active learning for forcefield parameterization, and generative modeling for amorphous materials.","UCLA  \nUCLA Electronic Theses and Dissertations  \nTitle  \nExpanding the frontiers of glass science: new insights derived from machine learning approaches  \nPermalink  \n[https://escholarship.org/uc/item/47k2s039](https://escholarship.org/uc/item/47k2s039)  \nAuthor  \nYang, Kai  \nPublication Date  \n2025  \nPeer reviewed|Thesis/dissertation  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nUNIVERSITY OF CALIFORNIA  \nLos Angeles  \nExpanding the frontiers of glass science:  \nnew insights derived from machine learning approaches  \nA dissertation submitted in partial satisfaction of the requirements for the degree Doctor of Philosophy in Civil Engineering  \nby  \nKai Yang  \n© Copyright by  \nKai Yang  \n2025  \nABSTRACT OF THE THESIS  \nExpanding the frontiers of glass science:  \nnew insights derived from machine learning approaches  \nby  \nKai Yang  \nDoctor of Philosophy in Civil Engineering  \nUniversity of California, Los Angeles, 2025  \nProfessor Fabian Rosner, Chair  \nGlasses, known as amorphous solids, are versatile materials with widespread applications across numerous fields (e.g., display application, optical fibers, laboratory equipment, nuclear waste immobilization, etc.) . Unlike crystalline solids, glasses possess greater compositional flexibility and are not constrained by strict stoichiometric requirements. This freedom from structural rules creates virtually unlimited possibilities for material exploration and design. Despite this vast compositional space, only a relatively small number of glass compositions have been synthesized throughout human history spanning thousands of years. This represents a significant  \nuntapped potential for discovering novel glass materials with enhanced properties and new functionalities.  \nComputational tools, particularly molecular dynamics (MD) simulations, can model the entire glass production process—from initial melting through the quenching stage—offering a complete virtual representation of glass synthesis. This computational approach bridges the gap between fundamental understanding and practical application, enabling researchers to explore vast compositional spaces efficiently before committing to experimental validation. Machine learning techniques, on the other hand, provide an efficient and accelerated pathway for designing glassy materials. The advancement of molecular dynamics simulations combined with machine learning methods enables a fully computational approach that can reveal the fundamental relationships between glass composition and properties. This integrated framework serves as a valuable guide for experimental design and manufacturing processes.  \nHere, I demonstrate how machine learning techniques can be integrated with computational methods to accelerate the discovery and development of novel glass compositions. Chapter one presents an analytical model for calculating topological constraints in calcium aluminosilicate (CAS) glass systems. Chapter two shows how these topological inputs can be leveraged to extrapolate the mechanical properties of CAS glasses. In Chapter three, I demonstrate how an active learning framework enables more efficient parameterization of classical molecular dynamics potentials, leading to improved forcefields for extensive glass systems. Finally, the last Chapter introduces a generative model for amorphous materials, showcasing how cutting-edge artificial intelligence can accelerate glass research and development.  \nThe dissertation of Kai Yang is approved.  \nDaniel Schwalbe-Koda  \nErtugrul Taciroglu  \nSanjey Mohanty  \nSriram Narasimhan  \nFabian Rosner, Committee Chair  \nUniversity of California, Los Angeles  \n2025  \nAcknowledgments  \nI would like to express my sincere gratitude to all those who have contributed to my academic journey and made this dissertation possible.  \nFirst and foremost, I am deeply grateful to my parents, wife, and family members for their unwavering support thro","cbCaioETK2UcuOkq","https://ap.wps.com/l/cbCaioETK2UcuOkq","pdf",10560375,1,137,"English","en",105,"# Abstract\n## Research motivation and approach\n## Chapter contributions\n## Acknowledgments","[{\"question\":\"Why is glass research considered promising despite limited historical synthesis?\",\"answer\":\"Glasses allow wide compositional freedom compared with crystalline stoichiometric limits, creating a large potential design space. However, only a small fraction of possible compositions has been synthesized, leaving room for novel materials with better properties and new functions.\"},{\"question\":\"How do molecular dynamics simulations contribute to the proposed computational framework?\",\"answer\":\"Molecular dynamics can model the full glass production pathway from melting through quenching, providing a virtual representation of glass synthesis. This helps connect fundamental mechanisms with practical property design before experiments.\"},{\"question\":\"What role does machine learning play in accelerating glass discovery in this dissertation?\",\"answer\":\"Machine learning is used to efficiently design and parameterize glassy materials by learning relationships between composition and properties. The dissertation demonstrates integration with analytical modeling, property extrapolation, active learning for forcefield improvements, and generative modeling for amorphous materials.\"}]","Expanding the frontiers of glass science: new insights derived from machine learning approaches - Doctor of Philosophy thesis | PDF",1785810131,345,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":28},"expanding-the-frontiers-of-glass-science-new-insights-derived-from-machine-learning-approaches-doctor-of-philosophy-thesis","",{"@graph":36,"@context":85},[37,54,68],{"@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/expanding-the-frontiers-of-glass-science-new-insights-derived-from-machine-learning-approaches-doctor-of-philosophy-thesis/122344/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is glass research considered promising despite limited historical synthesis?","Question",{"text":75,"@type":76},"Glasses allow wide compositional freedom compared with crystalline stoichiometric limits, creating a large potential design space. However, only a small fraction of possible compositions has been synthesized, leaving room for novel materials with better properties and new functions.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How do molecular dynamics simulations contribute to the proposed computational framework?",{"text":80,"@type":76},"Molecular dynamics can model the full glass production pathway from melting through quenching, providing a virtual representation of glass synthesis. This helps connect fundamental mechanisms with practical property design before experiments.",{"name":82,"@type":73,"acceptedAnswer":83},"What role does machine learning play in accelerating glass discovery in this dissertation?",{"text":84,"@type":76},"Machine learning is used to efficiently design and parameterize glassy materials by learning relationships between composition and properties. The dissertation demonstrates integration with analytical modeling, property extrapolation, active learning for forcefield improvements, and generative modeling for amorphous materials.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]