[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119934-en":3,"doc-seo-119934-105":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":20,"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},119934,4810365810221,"Aurora","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Interactive Machine Learning with Heterogeneous Data - Part","Interactive Machine Learning with Heterogeneous Data presents a dissertation on learning under interactive settings where feedback is heterogeneous across multiple tasks or agents. The work develops bandit methods that aggregate auxiliary and feedback sources, analyzes regret behavior, and establishes lower and upper bounds under known and unknown problem parameters. It further introduces Thompson-sampling-based robust transfer for multi-task bandits and extends the ideas to multi-task reinforcement learning using model transfer, followed by metric learning from crowdsourced preference comparisons.","UC San Diego  \nUC San Diego Electronic Theses and Dissertations  \nTitle  \nInteractive Machine Learning with Heterogeneous Data  \nPermalink  \n[https://escholarship.org/uc/item/9hh4c8fz](https://escholarship.org/uc/item/9hh4c8fz)  \nAuthor  \nWang, Zhi  \nPublication Date  \n2024  \nPeer reviewed|Thesis/dissertation  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nUNIVERSITY OF CALIFORNIA SAN DIEGO  \nInteractive Machine Learning with Heterogeneous Data  \nA dissertation submitted in partial satisfaction of the requirements for the degree Doctor of Philosophy  \nin  \nComputer Science  \nby  \nZhi Wang  \nCommittee in charge:  \nProfessor Kamalika Chaudhuri, Chair  \nProfessor Sanjoy Dasgupta  \nProfessor Tara Javidi  \nProfessor Yian Ma  \nCopyright Zhi Wang, 2024 All rights reserved.  \nThe Dissertation of Zhi Wang is approved, and it is acceptable in quality and form for publication on microfilm and electronically.  \nUniversity of California San Diego 2024  \nDEDICATION  \nTo my parents  \nEPIGRAPH  \n我想：希望本是无所谓有，无所谓无的。这正如地上的路；  \n其实地上本没有路，走的人多了，也便成了路。  \n— 鲁迅  \nI think: one cannot say that hope exists, or does not exist.  \nIt is like the roads that mark the earth; For though there were none at first, where footsteps have fallen, roads emerge.  \n—Lu Xun  \nTABLE OF CONTENTS  \nDissertation Approval Page ................................................ iii  \nDedication .............................................................. iv  \nEpigraph ................................................................ v  \nTable [of Contents ........................................................ vi](of Contents ........................................................ vi)  \n[List of Figures ........................................................... x](List of Figures ........................................................... x)  \n[Acknowledgements ....................................................... xii](Acknowledgements ....................................................... xii)  \n[Vita .................................................................... xiv](Vita .................................................................... xiv)  \n[Abstract of the Dissertation ............................................... xv](Abstract of the Dissertation ............................................... xv)  \n[Chapter 1 Introduction .................................................. 1](Chapter 1 Introduction .................................................. 1)  \n[Chapter 2 Multi-Task Bandits through Heterogeneous Feedback Aggregation ... 4](Chapter 2 Multi-Task Bandits through Heterogeneous Feedback Aggregation ... 4)  \n[2.1 Introduction ..................................................... 4](2.1 Introduction ..................................................... 4)  \n[2.2 The](2.2 The) ϵ-MPMAB Problem ........................................... 6  \n2.2.1 Can auxiliary data always help? .............................. 9  \n2.3 ϵ-MPMAB with Known ϵ .......................................... 10  \n2.3.1 Algorithm: RobustAgg (ϵ) .................................. 10  \n2.3.2 Regret analysis ............................................ 12  \n2.3.3 Lower bounds .............................................. 15  \n2.4 ϵ-MPMAB with Unknown ϵ ........................................ 16  \n2.4.1 Gap-dependent lower bound ................................. 16  \n2.4.2 Gap-independent upper bound ............................... 16  \n2.5 Related Work .................................................... 17  \n2.5.1 Multi-agent bandits ........................................ 17  \n2.5.2 Bandits in metric spaces .................................... 19  \n2.5.3 Learning using weighted data aggregation ..................... 19  \n2.6 Empirical Validation .............................................. 20  \n2.6.1 Experimental setup ......................................... 20  \n2.6.2 Simulations and results ..........................","cbCaiojVvZl9kUGW","https://ap.wps.com/l/cbCaiojVvZl9kUGW","pdf",19422460,1,362,"English","en",105,"# Chapter 1 Introduction\n# Chapter 2 Multi-Task Bandits through Heterogeneous Feedback Aggregation\n## 2.1 Introduction\n## 2.2 The ϵ-MPMAB Problem\n## 2.3 ϵ-MPMAB with Known ϵ\n## 2.4 ϵ-MPMAB with Unknown ϵ\n## 2.5 Related Work\n## 2.6 Empirical Validation\n## 2.7 Conclusion and Future Work\n# Chapter 3 Thompson Sampling for Robust Transfer in Multi-Task Bandits\n## 3.1 Introduction\n## 3.2 Preliminaries\n## 3.3 Algorithm: RobustAgg-TS (ϵ)\n## 3.4 Main Results\n## 3.5 Proof Ingredients\n## 3.6 Related Work\n## 3.7 Empirical Evaluation\n## 3.8 Conclusion","[{\"question\":\"What problem does the dissertation address?\",\"answer\":\"It focuses on interactive learning with heterogeneous data, especially when feedback differs across multiple tasks or agents. The research targets multi-task bandit and reinforcement learning settings and how to leverage auxiliary/heterogeneous feedback effectively.\"},{\"question\":\"How does the dissertation handle known versus unknown parameters in the multi-task bandit framework?\",\"answer\":\"Chapter 2 studies ϵ-MPMAB under both known ϵ and unknown ϵ. It provides regret analysis, robust aggregation algorithms, and corresponding upper and lower bounds.\"},{\"question\":\"What role does robust transfer and Thompson sampling play?\",\"answer\":\"Chapter 3 introduces a Thompson-sampling approach (RobustAgg-TS) designed for robust transfer in multi-task bandits. It includes main results, proof ingredients, and empirical evaluation.\"}]","Interactive Machine Learning with Heterogeneous Data - Part | PDF",1785727066,912,{"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},"interactive-machine-learning-with-heterogeneous-data-part","",{"@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/interactive-machine-learning-with-heterogeneous-data-part/119934/",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","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the dissertation address?","Question",{"text":75,"@type":76},"It focuses on interactive learning with heterogeneous data, especially when feedback differs across multiple tasks or agents. The research targets multi-task bandit and reinforcement learning settings and how to leverage auxiliary/heterogeneous feedback effectively.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the dissertation handle known versus unknown parameters in the multi-task bandit framework?",{"text":80,"@type":76},"Chapter 2 studies ϵ-MPMAB under both known ϵ and unknown ϵ. It provides regret analysis, robust aggregation algorithms, and corresponding upper and lower bounds.",{"name":82,"@type":73,"acceptedAnswer":83},"What role does robust transfer and Thompson sampling play?",{"text":84,"@type":76},"Chapter 3 introduces a Thompson-sampling approach (RobustAgg-TS) designed for robust transfer in multi-task bandits. It includes main results, proof ingredients, and empirical evaluation.","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"]