[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117563-en":3,"doc-seo-117563-105":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},117563,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Data Quality in Causal Machine Learning with Applications to Algorithmic Fairness","This thesis investigates data quality challenges in causal machine learning and develops methodological tools that connect causal identification to practical data limitations. The work focuses on how unmeasured confounding, hidden variables, selection bias, and multi-environment data affect recoverability of both observational distributions and causal effects. It further studies algorithmic fairness through a causal lens, contrasting non-causal fairness methods with causal fairness approaches to clarify when and how fairness can be justified from underlying causal structure.","Data Quality in Causal Machine Learning with Applications to Algorithmic Fairness  \nJake Howarth Fawkes  \nSt Hugh’s College University of Oxford  \nA thesis submitted for the degree of Doctor of Philosophy  \nHilary 2025  \nStatement of Originality  \nI hereby declare that except where specific reference is made to the work of others, the content of this thesis is my own work and has not been submitted in whole or in parts for any other degree or qualification. This thesis is my own work unless otherwise stated in the authorship form at the end of the chapters.  \nJake Fawkes Hillary 2025  \nI dedicate this thesis to Hilary for patiently listening to my ramblings and constantly supporting me throughout the years this took to write. In my opinion, you are the  \nbest of the world.  \nAcknowledgements  \nI want to begin by giving my deepest thanks to my supervisors, Robin J. Evans and Dino Sejdinovic. Robin, I am very grateful to you for believing in me and teaching me all the way from my first statistics and probability courses to the end of a PhD. I have learned so much under your guidance and I really appreciate the patience you had forme, especially given my lack of organisation. During my PhD, I always felt I had the right balance of intellectual freedom and support, and I am grateful to you for that. Dino, thank you for showing me first hand how much fun research can be. Working with you is a great reminder of how much passion and enthusiasm you can have for Math and ML. It is a great example for all who are lucky enough to work with you. I strongly hope I have the chance to cross paths with you both again in the future.  \nI would like to thank others who have guided me through projects over the years, specifically Zachary C. Litpon, Amartya Sanyal, Chris Holmes, and Uri Shalit. You were all much too generous with your time and thoughts; it was a privilege to learn from you. I also want to thank the groups who were kind enough to host me at various points over my PhD. Thank you so much to Krikamol Muandet for welcoming me in Saarbrücken and being a shining example of academic curiosity. Thank you to Ali Shah, Robert Grout and the rest of the team at Accenture for giving me my first internship. Thank you to Ciarán M. Gilligan-Lee, Michael O’ Riordan, Thanos Vlontzos, and Oriol Corcoll for supervising me during my time at Spotify’s advanced causal inference lab. You all created an incredibly friendly environment which was a pleasure to research in. Thank you to Jason Harford, Kristina Ulicna, and the rest of Valence labs for hosting me as an intern. Working with you all reinjected me with a passion for research that was dimming at the end of my PhD.  \nNext and very importantly, Coucou, Shasha. Meeting, working, and laughing with you was one of the true highlights of my PhD. I am still trying to figure out a way to get you to work with me again so that I can spend more of my days talking to you. I want to say a big thank you to Nic Fishman for being a good friend and an incredible collaborator. I always greatly enjoyed hearing your thoughts on the field and discussing  \nthings with you. Lucile, thank you very much for teaching me the Lucile method of being such a fun collaborator that people are desperate to work with you. Thank you to all the others who I have been lucky enough to work with over my PhD, specifically Robert, Desi, Omri, and Mel. It was a pleasure working with you all. Finally, thankyou to the rest of Dino’s group in Jef, Alan, and Veit for the many interesting ideas you all introduced me to over the PhD.  \nI was fortunate to have a number of great friendships throughout my PhD, which made the many years it took to complete much easier. Thank you to Patrick, Joris, Florence, JT, and the many others I was lucky enough to meet for being constant sources of fun and friendship away from study. Thank you to the other members of the best office in the stats department for making coming to work so much fun, specifically Eugen","cbCair3N1k5OAEgy","https://ap.wps.com/l/cbCair3N1k5OAEgy","pdf",4847142,1,194,"English","en",105,"# Introduction\n## Background on Causality\n## Causal Approaches to Data Quality\n## Algorithmic Fairness","[{\"question\":\"How does the thesis connect data quality to causal machine learning?\",\"answer\":\"It analyzes how common data-quality issues—like unmeasured confounding, hidden variables, and selection bias—impact the ability to identify and recover causal effects in causal machine learning workflows.\"},{\"question\":\"What causal assumptions are discussed when hidden variables are present?\",\"answer\":\"The thesis discusses how hidden variables can be handled via marginalisation and how identifiability of causal effects can change depending on what is assumed and observable.\"},{\"question\":\"How are algorithmic fairness approaches treated in the thesis?\",\"answer\":\"It presents both non-causal fairness methods and causal fairness methods, aiming to clarify how fairness claims can be supported using causal structure rather than only statistical correlations.\"}]","Data Quality in Causal Machine Learning with Applications to Algorithmic Fairness | PDF",1785677022,489,{"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},"data-quality-in-causal-machine-learning-with-applications-to-algorithmic-fairness","",{"@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/data-quality-in-causal-machine-learning-with-applications-to-algorithmic-fairness/117563/",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","2026-08-02",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},"How does the thesis connect data quality to causal machine learning?","Question",{"text":75,"@type":76},"It analyzes how common data-quality issues—like unmeasured confounding, hidden variables, and selection bias—impact the ability to identify and recover causal effects in causal machine learning workflows.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What causal assumptions are discussed when hidden variables are present?",{"text":80,"@type":76},"The thesis discusses how hidden variables can be handled via marginalisation and how identifiability of causal effects can change depending on what is assumed and observable.",{"name":82,"@type":73,"acceptedAnswer":83},"How are algorithmic fairness approaches treated in the thesis?",{"text":84,"@type":76},"It presents both non-causal fairness methods and causal fairness methods, aiming to clarify how fairness claims can be supported using causal structure rather than only statistical correlations.","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"]