[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122589-en":3,"doc-seo-122589-105":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},122589,7971461741311,"Ophelia","https://ap-avatar.wpscdn.com/avatar/74000253aff267980c6?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779345379180704826",8,"Research & Report","Advances in Kernel Methods - Towards General-Purpose and Scalable Models - Doctoral Dissertation","A wide range of statistical and machine learning tasks learn latent functions or their properties from data, including regression, classification, principal component analysis, optimization, point-process intensity learning, and reinforcement learning. Positive semi-definite kernels enable flexible nonparametric hypothesis spaces, yet recent breakthroughs often emphasize deep neural networks. This thesis develops theoretical and methodological foundations for fully automated, scalable, general-purpose kernel machines, bridging kernel methods and deep learning while aiming for strong performance on both small and large-scale problems, with contributions structured across two parts.","Advances in Kernel Methods  \nTowards General-Purpose and Scalable Models  \nYves-Laurent Kom Samo  \nDepartment of Engineering Science  \nUniversity of Oxford  \nThis dissertation is submitted for the degree of Doctor of Philosophy  \nSt. Anne’s College June 2017  \nTo my late father whom, 20 years on, I miss more than ever.  \nDeclaration  \nI hereby declare that except where specific reference is made to the work of others, the contents of this dissertation are original and have not been submitted in whole or in part for consideration for any other degree or qualification in this, or any other university. Notwithstanding the use of the first person of plural to express personal views throughout this thesis, this dissertation is my own work and contains nothing which is the outcome of work done in collaboration with others, except as specified in the text and Acknowledgements.  \nYves-Laurent Kom Samo June 2017  \nAcknowledgements  \nFirst and foremost, I would like to thank the Oxford-Man Institute (OMI) for the relentless support I have received throughout the course of my DPhil, without which this thesis would likely not have been possible. In particular, I would like to extend profuse thanks to Terry Lyons who was the Director of the OMI at the time of my admission, and with whom I have had numerous intellectually stimulating and very enjoyable discussions at the intersection of pure mathematics, data science and philosophy. My stay at the OMI wouldn’t have been the same without Marek Musiela, whose door is always open to any OMI students who could benefit from his immense wisdom of financial markets and his world-renowned expertise in stochastic analysis. Personally, I have been lucky to have many very fruitful discussions with Marek on stochastic analysis, finance and entrepreneurship, some of which have influenced my research and other personal endeavours. I would like to thank the great staff of the OMI, in particular Justin Sharp, Laura Okoli, and their teams, who have always warmly welcomed all my questions and requests. The IT facilities Justin’s team manage are nothing but first class, and Laura’s team have been of tremendous help in conference travel arrangements among other things.  \nI would like to thank my advisor Stephen Roberts who made my DPhil experience special, by giving me the freedom to drive my personal research agenda. I would also like to thank Google for sharing my enthusiasm for the research agenda I pursue in this thesis by awarding me a Google Fellowship in Machine Learning.  \nI am thankful for the love and camaraderie of my colleagues and friends. I will not attempt to name them all here for I would most likely forget important individuals. To all of you who have always genuinely been there for me I extend my sincere gratitude. I would like to extend special thanks to Jake Effoduh. His dynamism, kindness, indelible smile and indefectible commitment to be the voice of the unheard wherever he is have often swamped me in positivity, which undoubtedly has helped me in the course of my DPhil.  \nFinally, but most importantly, I would like to extend my heartfelt thanks to my family. I owe most of who I am today to my mother. Her strength after my father’s  \npassing has been an inspiration to me in many ways. The abundant attention and care of my sister Patricia towards me have helped me through bumpy rides on numerous occasions. My thoughts also go to my younger cousins Hermann, Jordan and Emmanuel, whom I love more than brothers. Showing them that they can achieve any goals they put their minds to has been a motivation of mine for many years. Lastly, my thoughts go to my late father. I find great comfort in thinking that this thesis would have made him proud, which is perhaps my ultimate motivation in life.  \nAbstract  \nA wide range of statistical and machine learning problems involve learning one or multiple latent functions, or properties thereof, from datasets. Examples include regression, classification, pr","cbCaimpknKykMvoM","https://ap.wps.com/l/cbCaimpknKykMvoM","pdf",21880144,1,348,"English","en",105,"# Abstract\n## Part I: Introduction and research gap\n## Part II: Flexible and scalable Bayesian kernel methods\n### Chapter 2: Inhomogeneous point process intensity inference\n### Chapter 3: Online forecasting and fully-online learning","[{\"question\":\"What kinds of machine learning problems does the thesis focus on?\",\"answer\":\"It targets problems that learn one or more latent functions (or properties) from datasets, such as regression, classification, principal component analysis, optimization, point-process intensity learning, and reinforcement learning.\"},{\"question\":\"How does the thesis position kernel methods relative to deep learning?\",\"answer\":\"It develops kernel-machine foundations intended to bridge kernel methods and deep learning, emphasizing scalable general-purpose performance on both small and large-scale settings, while highlighting potential advantages over deep learning.\"},{\"question\":\"What is the core direction of the thesis in Part II?\",\"answer\":\"Part II develops flexible and scalable Bayesian kernel methods aimed at datasets with locally homogeneous patterns, starting with applications that demonstrate how mild asymmetry in dependency structure can improve scalability, flexibility, and accuracy.\"}]","Advances in Kernel Methods - Towards General-Purpose and Scalable Models - Doctoral Dissertation | PDF",1785811574,877,{"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},"advances-in-kernel-methods-towards-general-purpose-and-scalable-models-doctoral-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/advances-in-kernel-methods-towards-general-purpose-and-scalable-models-doctoral-dissertation/122589/",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 kinds of machine learning problems does the thesis focus on?","Question",{"text":75,"@type":76},"It targets problems that learn one or more latent functions (or properties) from datasets, such as regression, classification, principal component analysis, optimization, point-process intensity learning, and reinforcement learning.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the thesis position kernel methods relative to deep learning?",{"text":80,"@type":76},"It develops kernel-machine foundations intended to bridge kernel methods and deep learning, emphasizing scalable general-purpose performance on both small and large-scale settings, while highlighting potential advantages over deep learning.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the core direction of the thesis in Part II?",{"text":84,"@type":76},"Part II develops flexible and scalable Bayesian kernel methods aimed at datasets with locally homogeneous patterns, starting with applications that demonstrate how mild asymmetry in dependency structure can improve scalability, flexibility, and accuracy.","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"]