[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-290662-105":59,"doc-detail-290662-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","hardware-acceleration-of-machine-learning-and-combinatorial-optimization-problems-a-dissertation-doctor-of-philosophy","Hardware Acceleration of Machine Learning and Combinatorial Optimization Problems - A Dissertation - Doctor of Philosophy","","The rapid advancement of artificial intelligence intensifies the demand for energy-efficient computing across both data centers and edge devices. Among AI workloads, machine learning and combinatorial optimization are two of the most computationally demanding yet fundamentally distinct domains. Machine learning features dense and parallelizable dataflows, while combinatorial optimization relies on irregular, constraint-driven computation. This dissertation investigates hardware acceleration for both domains using application-specific integrated circuits.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/hardware-acceleration-of-machine-learning-and-combinatorial-optimization-problems-a-dissertation-doctor-of-philosophy/290662/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/hardware-acceleration-of-machine-learning-and-combinatorial-optimization-problems-a-dissertation-doctor-of-philosophy/290662.png","ImageObject",300,407,{"name":92,"@type":93},"Levi","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-06","2026-09-17",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",14,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does the dissertation target?","Question",{"text":113,"@type":114},"It targets energy-efficient computing for artificial intelligence workloads on both data centers and edge devices, focusing on machine learning and combinatorial optimization.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does the dissertation distinguish between ML and CO workloads?",{"text":118,"@type":114},"ML is described as having dense and parallelizable dataflows, while CO involves irregular computations driven by constraints.",{"name":120,"@type":111,"acceptedAnswer":121},"What hardware approach is used for acceleration?",{"text":122,"@type":114},"The dissertation explores hardware acceleration using application-specific integrated circuits (ASICs) and presents a template-based framework for accelerating shallow ML algorithms.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},290662,1789642310,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":130,"read_time":145,"preview_page_count":4},7971461740909,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","Hardware Acceleration of Machine Learning and Combinatorial Optimization Problems  \nA DISSERTATION  \nSUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA  \nBY  \nZiqing Zeng  \nIN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  \nDoctor of Philosophy  \nSachin S. Sapatnekar, Advisor  \nJanuary, 2026  \n© Ziqing Zeng 2026 ALL RIGHTS RESERVED  \nAcknowledgements  \nT. S. Eliot once wrote,“We shall not cease from exploration, and the end of all our exploring will be to arrive where we started and know the place for the first time.” This reflection captures my feelings as I bring my Ph.D. work to a close. When I joined the VLSI EDA Lab at the University of Minnesota, Twin Cities in 2019, I had little sense of the journey ahead. Alongside the isolation imposed by the pandemic, the early stages of this exploration made the path feel endless—much like Minneapolis in deep winter, when the city seems perpetually frozen. Yet, as Eliot suggests, true understanding emerges through the journey itself. With the support of many people, my Ph.D. experience gradually evolved from an overwhelming challenge into an unforgettable and invaluable chapter of my life. It is with deep gratitude that I offer my thanks to those who helped, guided, and supported me throughout this journey.  \nI would like to begin by expressing my deepest gratitude to my advisor, Prof. Sachin S. Sapatnekar. It has been a true honor to pursue my doctoral studies under his supervision. In the early stages of my research, my approach was often rushed, focused more on immediate data than on the underlying logic of the broader context. I initially struggled to appreciate the meticulous standards required for academic writing and presentation. Under his guidance, sustained effort led to a fundamental transformation. I learned theart of the \"big-picture\" perspective—learning to analyze the fundamental nature of a problem, identify core roadblocks, and decompose complex challenges into a clear, navigable path forward. The lessons he instilled in me extend far beyond academic research, profoundly shaping my attitude toward work and life. It is through his mentorship that I have come to realize the Ph.D. journey is not only about research, but about developing a way of thinking—a true philosophy of life.  \nI would like to thank my committee members: Prof. Chris H. Kim, Prof. Ulya R.  \nKarpuzcu, and Prof. Anthonia Zhai for their insightful guidance on my dissertation. Over the years, I have had the privilege of collaborating with Prof. Kim and Prof. Karpuzcu on research projects. Their creative perspectives and steady support were precious influences helping me overcome research challenges. I also thank the Defense Advanced Research Projects Agency (DARPA) and the ECE Department at the University of Minnesota for providing the financial support and resources necessary to carryout my research.  \nI would like to thank my wonderful labmates: Dr. Meghna, Dr. Masoud, Dr. Tonmoy, Dr. Vidya, Dr. Kunal, Dr. Ramprasath, Dr. Sudipta, Dr. Nibedita, Dr. Wenjing, Dr. Shohel, Subhadip, Hangyu, Divya, Ziyang, Endalk, Soasina, Peiyi, and especially Dr. Susmita and Abhimanyu, who collaborated with me on projects. I am deeply grateful for the constant help and support I received from my labmates. I will always cherish the fun-filled moments and thoughtful discussions with this amazing group.  \nI would like to thank my collebrators in projects: Prof. Andrew B. Kahng, Dr. Zhiang and Sayak (from USCD); Dr. Will, Dr. Hao, Dr. Hüsrev, Chaohui, Ruihong, Xunqi, Peter, Ahmet, and Nafisa. It was a great privilege to work with and learn from them.  \nBeing far away from home, studying abroad is challenging. I would like to thank my friends: Dr. Yangyang, Dr. Hanzhao, Dr. Wen, Dr. Zhongqi, Tiancong, Yixuan, Shulang, Mengting, Wanting, Yujie, and Jiaying, who helped me endure the winters in Minnesota.  \nMost importantly, I would like to acknowledge the greatest source of my inspiratio","cbCaipLnlzawqYiM","https://ap.wps.com/l/cbCaipLnlzawqYiM","pdf",5591534,160,"English","# Acknowledgements\n## Advisor and research guidance\n## Committee support and funding\n## Labmates, collaborators, and friends\n## Family dedication\n# Dedication\n# Abstract\n## Motivation and workload distinctions\n## Proposed ML acceleration framework","[{\"question\":\"What problem does the dissertation target?\",\"answer\":\"It targets energy-efficient computing for artificial intelligence workloads on both data centers and edge devices, focusing on machine learning and combinatorial optimization.\"},{\"question\":\"How does the dissertation distinguish between ML and CO workloads?\",\"answer\":\"ML is described as having dense and parallelizable dataflows, while CO involves irregular computations driven by constraints.\"},{\"question\":\"What hardware approach is used for acceleration?\",\"answer\":\"The dissertation explores hardware acceleration using application-specific integrated circuits (ASICs) and presents a template-based framework for accelerating shallow ML algorithms.\"}]","Hardware Acceleration of Machine Learning and Combinatorial Optimization Problems - A Dissertation - Doctor of Philosophy | PDF",403]