[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126922-en":3,"doc-seo-126922-105":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},126922,2336474459895,"Aria","https://ap-avatar.wpscdn.com/avatar/22000baeef7a5ed0655?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786071322749376916",8,"Research & Report","Optimization Algorithms for Machine Learning - Dissertation - 机器学习优化算法论文","This dissertation presents optimization algorithms for machine learning, focusing on coordinate descent and stochastic gradient descent in mini-batch settings. It clarifies the research objectives through a set of major questions partially or fully answered in the work, and details specific contributions across a dedicated chapter. Supporting material is provided via multiple appended manuscripts covering screening rules, approximate steepest coordinate descent, safe adaptive importance sampling, matching pursuit with coordinate descent, and variance reduction methods for large-scale optimization.","Optimization Algorithms for Machine Learning  \nDissertation  \nder Mathematisch-Naturwissenschaftlichen Fakultät der Eberhard Karls Universität Tübingen zur Erlangung des Grades eines Doktors der Naturwissenschaften  \n(Dr. rer. nat. )  \nvorgelegt von  \nAnant Raj  \naus Lakhisarai, India  \nTübingen  \n2020  \nGedruckt mit Genehmigung der Mathematisch-Naturwissenschaftlichen Fakultät der Eberhard Karls Universität Tübingen.  \nTag der mündlichen Qualifikation: 07.06.2021  \nStellvertretender Dekan: Prof. Dr. József Fortágh  \n1. Berichterstatter/-in: Prof. Bernhard Schölkopf  \n2. Berichterstatter/-in: Prof. Philipp Hennig  \nTo my Parents and my sisters Anisha andAmrita.  \n“In exactly the same way, ... scatter your body, your feeling, your perception, your predispositions, your discriminative consciousness, break them up, knock them down, cease to play with them, apply yourself to the destruction of craving for them. Verily, ... the extinction of craving is Nirvana.”  \nBuddha  \n“All things appear and disappear because of the concurrence of causes and conditions. Nothing ever exists entirely alone; everything is in relation to everything else” -Buddha Image Source: [pixelbay.com](pixelbay.com)  \nPreface  \nThe significant portion of the work presented in this dissertation has been carried out between August 2015 and August 2020 at the Max-Planck institute for Intelligent Systems, Tübingen in the Empirical Inference Department headed by Dr. Bernhard Schölkopf. A small but important portion of the work included in this dissertation has been done while I was visiting the group led by Prof. Martin Jaggi at EPFL, Lausanne and by Prof. Francis Bach at SIERRA-Inria, Paris to do research in the theory of convex/non-convex optimization for machine learning.  \nThis thesis is organized overall in 11 chapters out of which there are 3 main chapters which contain introduction, objective and contribution made in this thesis. 8 of the remaining chapters are included in the appendix which have been taken directly from my research manuscripts.  \n• Chapter 1 is a brief introduction of the topic and provides background on coordinate descent and stochastic gradient descent optimization methods. It also contains a discussion on the major research challenges in mini batch stochastic gradient descent and coordinate descent algorithms.  \n• Chapter 2 briefly discusses the objective of this thesis. It contains a concise description of the major research questions which have been partially/fully answered in this thesis.  \n• Chapter 3 contains the specific contributions made in thesis. It further mentions all the research manuscripts included in this thesis and describes the contribution made by me in each manuscript. In the final part of this chapter starting from Chapter 3.4, I discuss the background and main results of each manuscript in a concise manner.  \n• Appendix A contains the copy of manuscript titled “Screening Rules for Convex Problems” which has been presented in Optimization for Machine Learning Worksop at Neurips 2016, held in Barcelona.  \n• Appendix B contains the copy of the manuscript titled “Approximate Steepest Coordinate Descent” published at ICML, 2017 held in Sydney, Australia.  \n• Appendix C contains the copy of the manuscript titled “Safe Adaptive Importance Sampling” published at Neurips, 2017 held in USA.  \n• Appendix D contains the copy of the manuscript titled “On Matching Pursuit and Coordinate Descent” published at ICML, 2018 held in Stockholm, Sweden.  \n• Appendix E contains the copy of the manuscript titled “k-SVRG: Variance Reduction for Large Scale Optimization” which is an arXiv Manuscript.  \n• Appendix F contains the copy of the manuscript titled “A Simpler Approach to Accelerated Stochastic Optimization” published at ICML, 2020 held online.  \n• Appendix G contains the copy of the manuscript titled “Importance Sampling via Local Sensitivity” published at AISTATS, 2020 held online.  \n• Appendix H contains the copy of the manuscript tit","cbCailIucUYqhkLk","https://ap.wps.com/l/cbCailIucUYqhkLk","pdf",15355622,1,348,"English","en",105,"# Preface\n## Chapter 1: Introduction\n## Chapter 2: Objective\n## Chapter 3: Contributions\n## Appendix A\n## Appendix B\n## Appendix C\n## Appendix D\n## Appendix E\n## Appendix F\n## Appendix G\n## Appendix H\n# Acknowledgements","[{\"question\":\"What optimization methods are highlighted in the introduction chapter?\",\"answer\":\"The dissertation introduction covers coordinate descent and stochastic gradient descent, including challenges in mini-batch stochastic gradient descent and coordinate descent algorithms.\"},{\"question\":\"How does the thesis define its objective?\",\"answer\":\"Chapter 2 provides a concise description of the major research questions that are partially or fully answered in the dissertation.\"},{\"question\":\"What kinds of materials are included in the appendix chapters?\",\"answer\":\"The appendix contains copies of multiple research manuscripts, each focusing on a specific contribution such as screening rules, approximate steepest coordinate descent, and importance sampling variants.\"}]","Optimization Algorithms for Machine Learning - Dissertation - 机器学习优化算法论文 | PDF",1785935671,877,{"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},"optimization-algorithms-for-machine-learning-dissertation-machine-learning-optimization-algorithms-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/optimization-algorithms-for-machine-learning-dissertation-machine-learning-optimization-algorithms-thesis/126922/",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","2026-08-05",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 optimization methods are highlighted in the introduction chapter?","Question",{"text":75,"@type":76},"The dissertation introduction covers coordinate descent and stochastic gradient descent, including challenges in mini-batch stochastic gradient descent and coordinate descent algorithms.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the thesis define its objective?",{"text":80,"@type":76},"Chapter 2 provides a concise description of the major research questions that are partially or fully answered in the dissertation.",{"name":82,"@type":73,"acceptedAnswer":83},"What kinds of materials are included in the appendix chapters?",{"text":84,"@type":76},"The appendix contains copies of multiple research manuscripts, each focusing on a specific contribution such as screening rules, approximate steepest coordinate descent, and importance sampling variants.","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"]