[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124394-en":3,"doc-seo-124394-105":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":4,"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},124394,1374391974564,"Clementine","https://ap-avatar.wpscdn.com/avatar/14000253aa45c000a9e?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779874745381141002",8,"Research & Report","Architecture Optimization and Data-Efficient Methods in Machine Learning","Yu Wang’s dissertation addresses how to optimize machine-learning architectures while improving data efficiency. The work frames efficient learning as both an algorithmic and systems problem, combining research prototypes and accelerator-oriented contributions. It builds on multiple projects including NUTS accelerator, AutoNF, TS-NODE, and ADO-LLM, and situates the methods within electrical and computer engineering research practice. Funding acknowledgements highlight support from the U.S. Department of Energy and the National Science Foundation.","UC Santa Barbara  \nUC Santa Barbara Electronic Theses and Dissertations  \nTitle  \nArchitecture Optimization and Data-Efficient Methods in Machine Learning  \nPermalink  \n[https://escholarship.org/uc/item/4rh0v7fg](https://escholarship.org/uc/item/4rh0v7fg)  \nISBN  \n9798297663756  \nAuthor  \nWang, Yu  \nPublication Date  \n2025-09-12  \nPeer reviewed|Thesis/dissertation  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nUniversity of California  \nSanta Barbara  \nArchitecture Optimization and Data-Efficient Methods in Machine Learning  \nA dissertation submitted in partial satisfaction  \nof the requirements for the degree  \nDoctor of Philosophy  \nin  \nElectrical and Computer Engineering  \nby  \nYu Wang  \nCommittee in charge:  \nProfessor Peng Li, Chair  \nProfessor Kerem Y C¸amsarı  \nProfessor Upamanyu Madhow  \nProfessor Yao Qin  \nSeptember 2025  \nThe Dissertation of Yu Wang is approved.  \n\n| Professor Kerem Y C¸amsarı |\n| --- |\n| Professor Upamanyu Madhow |\n| Professor Yao Qin |\n\nProfessor Peng Li, Committee Chair  \nJuly 2025  \nArchitecture Optimization and Data-Efficient Methods in Machine Learning  \nCopyright © 2025  \nby  \nYu Wang  \nTo my parents, for their love and patience  \nAcknowledgements  \nI would like to begin by expressing my deepest gratitude to my advisor, Prof. Peng Li. His guidance, patience, and constant encouragement have been the backbone of this dissertation. Beyond technical mentorship, his insistence on clarity of thought and rigor in research has shaped not only the work presented here but also how I will approach challenges in the future. I am also thankful to my committee members, Prof. Yao Qin, Prof. Kerem C¸amsarı, and Prof. Upamanyu Madhow, for their invaluable feedback, sharp insights, and willingness to generously share their time and expertise.  \nI am deeply grateful to my collaborators and labmates, whose insightful discussionsand spirited debates helped transform many half-formed sketches into the results presented in these pages. I owe particular thanks to my co-authors on the NUTS accelerator [1], AutoNF [2], TS-NODE [3], and ADO-LLM [4] projects for their dedication, patience with my endless requests for clarification and revision, and for reminding me that meaningful research depends not only on novel ideas, but equally on patience, persistence, and collaboration.  \nSpecifically, the aforementioned research projects were supported by the U.S. Department of Energy, Office of Advanced Scientific Computing Research, through the Data-Driven Decision Control for Complex Systems (DnC2S) project under Award DESC0021321, and by the National Science Foundation under Awards 1956313, 2334380, and 1741173 . I sincerely appreciate their generous support and investment.  \nMost importantly, I thank my family. Their unconditional love, belief in me, and endless patience have carried me through the inevitable ups and downs of doctoral life. To my parents: this work is dedicated to you. Without your support and encouragement, I would never have reached this point.  \nFinally, a special thanks to ChatGPT, Copilot and other tireless AI collaborators.  \nFrom polishing papers to prototyping research code, they have been uncredited labmates who never complained—except perhaps when I exceed the rate limit. While they cannot attend my defense, their fingerprints are quietly hidden in many of these pages.  \nCurriculum Vitæ  \nYu Wang  \nEducation  \n2025 Ph.D. in Computer Engineering (Expected), University of Califor  \nnia, Santa Barbara.  \n2019 M.S. in Computer Engineering, University of California, Santa Bar  \nbara.  \n2017 B.Eng in Microelectronics Science and Engineering. Fudan Univer  \nsity, Shanghai  \nPublications  \nYu Wang, Kamalika Das, Xiang Gao, Wendi Cui, Peng Li, Jiaxin Zhang, “Gradientguided Attention Map Editing: Towards Efficient Contextual Hallucination Mitigation”, Findings of The 2025 Annual Conference of the Nations of the Americas Chapter of the Associati","cbCainidb3HXVadr","https://ap.wps.com/l/cbCainidb3HXVadr","pdf",3626189,1,123,"English","en",105,"# Acknowledgements\n## Collaborators and funding support\n# Curriculum Vitæ\n## Education\n## Publications","[{\"question\":\"What is the dissertation’s main topic?\",\"answer\":\"The dissertation focuses on architecture optimization and data-efficient methods in machine learning, aiming to improve efficiency in learning systems.\"},{\"question\":\"Who provided guidance and feedback during the dissertation process?\",\"answer\":\"The advisor Prof. Peng Li is credited for guidance and rigor, and committee members including Prof. Yao Qin, Prof. Kerem Çamsarı, and Prof. Upamanyu Madhow are thanked for feedback.\"},{\"question\":\"Which projects are highlighted as key research work?\",\"answer\":\"The acknowledgements reference the NUTS accelerator, AutoNF, TS-NODE, and ADO-LLM projects as central collaborations contributing to results presented in the dissertation.\"}]","Architecture Optimization and Data-Efficient Methods in Machine Learning | PDF",1785821971,310,{"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},"architecture-optimization-and-data-efficient-methods-in-machine-learning","",{"@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/architecture-optimization-and-data-efficient-methods-in-machine-learning/124394/",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":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What is the dissertation’s main topic?","Question",{"text":75,"@type":76},"The dissertation focuses on architecture optimization and data-efficient methods in machine learning, aiming to improve efficiency in learning systems.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Who provided guidance and feedback during the dissertation process?",{"text":80,"@type":76},"The advisor Prof. Peng Li is credited for guidance and rigor, and committee members including Prof. Yao Qin, Prof. Kerem Çamsarı, and Prof. Upamanyu Madhow are thanked for feedback.",{"name":82,"@type":73,"acceptedAnswer":83},"Which projects are highlighted as key research work?",{"text":84,"@type":76},"The acknowledgements reference the NUTS accelerator, AutoNF, TS-NODE, and ADO-LLM projects as central collaborations contributing to results presented in the dissertation.","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"]