[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123349-en":3,"doc-seo-123349-105":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},123349,962075114765,"Quinn","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","User Agency Across the Machine Learning Pipeline - Thesis Summary","User Agency Across the Machine Learning Pipeline examines how algorithmic systems can shape, constrain, or enable user agency across end-to-end learning workflows. It focuses on algorithmic censoring in dynamic learning systems, proposing mechanisms and mitigation strategies involving exploration and recourse, and evaluating results through carefully defined experiments and metrics. The dissertation further addresses representational harms and their measurement, extending fairness considerations toward personalization, with an emphasis on epistemic utility and equity for modern models, including large language models.","UC San Diego  \nUC San Diego Electronic Theses and Dissertations  \nTitle  \nUser Agency Across the Machine Learning Pipeline  \nPermalink  \n[https://escholarship.org/uc/item/7br693r0](https://escholarship.org/uc/item/7br693r0)  \nAuthor  \nChien, Jennifer  \nPublication Date  \n2025  \nPeer reviewed|Thesis/dissertation  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nUNIVERSITY OF CALIFORNIA SAN DIEGO  \nUser Agency Across the Machine Learning Pipeline  \nA dissertation submitted in partial satisfaction of the requirements for the degree Doctor of Philosophy  \nin  \nComputer Science  \nby  \nJennifer Jiunyi Chien  \nCommittee in charge:  \nProfessor David Danks, Co-Chair  \nProfessor Margaret Roberts, Co-Chair  \nProfessor Lawrence Saul  \nProfessor Kristen Vaccaro  \nCopyright  \nJennifer Jiunyi Chien, 2025 All rights reserved.  \nThe Dissertation of Jennifer Jiunyi Chien is approved, and it is acceptable in quality and form for publication on microfilm and electronically.  \nUniversity of California San Diego  \n2025  \nEPIGRAPH  \nThis search for what you want is like tracking something  \nthat doesn’t want to be tracked. It takes time to get a dance right, to create something memorable.  \nFred Astaire  \nThis life is mine alone.  \nSo I have stopped asking people for directions  \nto places they’ve never been.  \nGlennon Doyle  \nWe have been taught to fear the very things that have the potential to set us free.  \nAlok Vaid-Menon  \nTABLE OF CONTENTS  \nDissertation Approval Page .................................................... iii  \nEpigraph .................................................................... iv  \nTable of Contents ............................................................ v  \nList of Figures ............................................................... vii  \nList of Tables ................................................................ viii  \nPreface ..................................................................... ix  \nAcknowledgements ........................................................... x  \nVita ........................................................................ xii  \nAbstract of the Dissertation .................................................... xiv  \nIntroduction ................................................................. 1  \n0.0.1 Background and Motivation ..................................... 1  \n0.0.2 Assumptions .................................................. 2  \n0.0.3 Research Objectives ............................................ 3  \n0.0.4 Thesis Structure ............................................... 5  \nChapter 1 Algorithmic Censoring in Dynamic Learning Systems .................. 7  \n1.1 Introduction ......................................................... 7  \n1.2 Framework .......................................................... 11  \n1.2.1 Preliminaries .................................................. 11  \n1.2.2 Censoring .................................................... 12  \n1.3 Mechanisms that Induce Censoring ...................................... 13  \n1.4 Mitigation ........................................................... 18  \n1.4.1 Exploration ................................................... 19  \n1.4.2 Recourse ..................................................... 22  \n1.5 Experiments ......................................................... 24  \n1.5.1 Setup ........................................................ 24  \n1.5.2 Results ....................................................... 26  \n1.6 Concluding Remarks .................................................. 31  \n1.7 Appendix: Additional Details on Experiments ............................. 33  \n1.7.1 Data Generating Techniques ..................................... 33  \n1.7.2 Data Generating Processes ...................................... 33  \n1.8 Metrics ............................................................. 36  \n1.9 Full Results Table .....","cbCaiqXsyYjkSDPT","https://ap.wps.com/l/cbCaiqXsyYjkSDPT","pdf",4450478,1,135,"English","en",105,"# Introduction\n## Background and Motivation\n## Assumptions\n## Research Objectives\n## Thesis Structure\n# Chapter 1 Algorithmic Censoring in Dynamic Learning Systems\n## Introduction\n## Framework\n## Mechanisms that Induce Censoring\n## Mitigation\n## Experiments\n## Concluding Remarks\n# Chapter 2 Beyond Behaviorist Representational Harms\n## Measurement and Mitigation\n## Representational Harms: Definition and Limitations\n## Recipes for Real-World Measurement\n## Additional Challenges with Large Language Models\n## Proposed Mitigations and Limitations\n## Conclusion\n# Chapter 3 Fairness Vs. Personalization\n## Towards Equity in Epistemic Utility","[{\"question\":\"What problem does the dissertation address about user agency?\",\"answer\":\"It studies how machine learning pipelines affect users’ ability to act and make choices across the learning process. The work analyzes constraints introduced by system behaviors and proposes ways to reduce harm and enable more effective agency.\"},{\"question\":\"How does the dissertation handle algorithmic censoring?\",\"answer\":\"It introduces mechanisms that induce censoring in dynamic learning systems and presents mitigation strategies. The proposed approaches include exploration and recourse, evaluated through experiments and supporting metrics.\"},{\"question\":\"How does the dissertation connect representational harms and fairness?\",\"answer\":\"It argues that representational harms require measurement beyond narrow behaviorist views. It then extends the discussion to fairness versus personalization by aiming for equity in epistemic utility.\"}]","User Agency Across the Machine Learning Pipeline - Thesis Summary | PDF",1785816067,340,{"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},"user-agency-across-the-machine-learning-pipeline-thesis-summary","",{"@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/user-agency-across-the-machine-learning-pipeline-thesis-summary/123349/",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 problem does the dissertation address about user agency?","Question",{"text":75,"@type":76},"It studies how machine learning pipelines affect users’ ability to act and make choices across the learning process. The work analyzes constraints introduced by system behaviors and proposes ways to reduce harm and enable more effective agency.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the dissertation handle algorithmic censoring?",{"text":80,"@type":76},"It introduces mechanisms that induce censoring in dynamic learning systems and presents mitigation strategies. The proposed approaches include exploration and recourse, evaluated through experiments and supporting metrics.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the dissertation connect representational harms and fairness?",{"text":84,"@type":76},"It argues that representational harms require measurement beyond narrow behaviorist views. It then extends the discussion to fairness versus personalization by aiming for equity in epistemic utility.","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"]