[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124514-en":3,"doc-seo-124514-105":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},124514,7971461741311,"Ophelia","https://ap-avatar.wpscdn.com/avatar/74000253aff267980c6?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779345379180704826",8,"Research & Report","Domain-Knowledge-Guided Machine Learning for Networked Systems - Dissertation","Advances in networked systems such as 5G networks and smart grids connect cyber and physical worlds and reshape daily life. Machine learning offers strong predictive power, yet applying black-box methods to networked systems is constrained by correlation, interdependence, and system rules. This dissertation develops domain-knowledge-guided ML that manages, models, and enhances networked systems by integrating domain knowledge and aligning models with constraints. Three research directions address physics-guided power flow, generative deep caching approximating optimal policies, and additional constraint-aware learning for networked environments.","Domain-Knowledge-Guided Machine Learning for  \nNetworked Systems  \nA DISSERTATION  \nSUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA  \nBY  \nXinyue Hu  \nIN PARTIAL FULFILLMENT OF THE REQUIREMENTS  \nFOR THE DEGREE OF  \nDOCTOR OF PHILOSOPHY  \nProfessor Zhi-Li Zhang  \nMarch, 2025  \n© Xinyue Hu 2025 ALL RIGHTS RESERVED  \nAcknowledgements  \nI am deeply grateful to many people for their contributions throughout my time in graduate school.  \nFirst and foremost, I would like to express my deepest appreciation to my advisor, Professor Zhi-Li Zhang, for his unwavering encouragement, generous support, constructive criticism, and invaluable guidance throughout my Ph.D. journey. He provided me with numerous opportunities to explore, learn, and grow, especially in the early stages of my PhD study. His inclusiveness allowed me to engage in a wide range of research projects, broadening my academic perspective and sharpening my research skills. Through his mentorship, I learned not only how to think critically but also how to systematically and iteratively design and conduct research. He instilled in me the importance of collaboration and helped me transform my hesitant speaking and writing into compelling, well-structured narratives. I have been continually inspired by his knowledge, dedication, and relentless pursuit of excellence. Beyond academia, his genuine care for students’well-being made a profound impact on my journey. Completing this dissertation while raising a child would not have been possible without his support, for which I am deeply grateful.  \nI am grateful to Professor Ali Anwar, Professor Jeff Calder, and Professor Vipin Kumar for taking time to serve on my thesis committee. I am especially thankful to Professor Vipin Kumar for his continued support throughout my Ph.D. journey, serving on my qualification, thesis proposal, and defense committees, as well as writing a recommendation letter for my dissertation fellowship application.  \nI had the privilege of working with and learning from many outstanding researchers and engineers. I would like to express my sincere gratitude to my collaborators: Wei Ye, Professor Eman Ramadan, Ziyan Wu, Jiaxiang Tang, Arvind Narayanan, Rishabh  \nMehta, Qixin Zhang, Saurabh Verma, Feng Tian, Rostand AK Fezeu, Professor Xin Liu, Professor Feng Qian, Professor Arnob Ghosh, and Professor Ness Shroff, for their time, support, and invaluable insights. Their feedback was instrumental in shaping my research and bringing it to fruition. I am also deeply grateful to Professor Xin Zhang and Professor Yanhua Li for generously sharing their machine learning expertise. My heartfelt appreciation goes to my internship mentors, Zihui Ge and Manjula Shivanna, for their invaluable guidance and support. Furthermore, I would like to extend my sincere thanks to Steven Sleder, Jason Carpenter, and Anlan Zhang for their assistance with my experiment setups. Lastly, I am especially thankful to our department staff members Joseph Nieszner, Phil Croteau, and Mary G. Nissen for their support in facilitating my parental leave during my Ph.D. studies. Their help made a significant difference in my journey.  \nLast but not least, I would like to express my love to my son, Xujie Duan, whose presence inspires me to grow and strengthens my determination to pursue a Ph.D. , setting an example for him. I am profoundly grateful to my husband, parents, and brother for their unwavering support and encouragement, which have allowed me to follow my passions.  \nMy research was supported by various sources of funds: NSF Grants CNS-1814322, CNS-1831140, CNS-1901103, and CNS-1952085, CNS-1836772, CCF-2123987, CNS- 2106771, CNS-2128489, CNS-2312836, CNS-2106933, CNS-2112471, and CNS-1955535, US DoD DTRA Grant HDTRA1-14-1-0040, and an Amazon AWS ML Research Award.  \nDedication  \nTo my son, who is my heart living outside of my body.  \nAbstract  \nAdvances in networked systems, such as 5G networks and smart grids,","cbCaifEmaqEtxKv3","https://ap.wps.com/l/cbCaifEmaqEtxKv3","pdf",9837360,1,152,"English","en",105,"# Abstract\n## Aims and motivation\n## Physics-guided modeling for power grids\n## Constraint-aligned ML for caching policies\n## Research contributions","[{\"question\":\"为什么将黑盒机器学习直接应用到网络化系统会遇到困难？\",\"answer\":\"网络化系统具有内在的相关性、相互依赖性以及遵循的规则和约束。黑盒方法往往忽略这些特性，可能导致设计失败或出现错误设计。\"},{\"question\":\"论文如何在电力网建模中融合领域知识与机器学习？\",\"answer\":\"将电力系统的物理定律（如基尔霍夫定律）与系统拓扑引入神经网络，用于在系统模型不准确或不可得时求解功率潮流问题。物理引导神经网络在精度与泛化性上优于不受约束的数据驱动方法。\"},{\"question\":\"论文在缓存策略方面提出了什么方法？\",\"answer\":\"将 Belady 的 MIN 理论见解与生成式神经网络的预测能力结合，构建深度学习缓存算法，用于逼近理论最优策略，从而降低时延并提升带宽利用效率。\"}]","Domain-Knowledge-Guided Machine Learning for Networked Systems - Dissertation | PDF",1785822850,383,{"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},"domain-knowledge-guided-machine-learning-for-networked-systems-dissertation","",{"@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/domain-knowledge-guided-machine-learning-for-networked-systems-dissertation/124514/",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},"为什么将黑盒机器学习直接应用到网络化系统会遇到困难？","Question",{"text":75,"@type":76},"网络化系统具有内在的相关性、相互依赖性以及遵循的规则和约束。黑盒方法往往忽略这些特性，可能导致设计失败或出现错误设计。","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"论文如何在电力网建模中融合领域知识与机器学习？",{"text":80,"@type":76},"将电力系统的物理定律（如基尔霍夫定律）与系统拓扑引入神经网络，用于在系统模型不准确或不可得时求解功率潮流问题。物理引导神经网络在精度与泛化性上优于不受约束的数据驱动方法。",{"name":82,"@type":73,"acceptedAnswer":83},"论文在缓存策略方面提出了什么方法？",{"text":84,"@type":76},"将 Belady 的 MIN 理论见解与生成式神经网络的预测能力结合，构建深度学习缓存算法，用于逼近理论最优策略，从而降低时延并提升带宽利用效率。","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"]