[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117149-en":3,"doc-seo-117149-105":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},117149,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","Structural and Statistical Uncertainty in Observational Causal Machine Learning at Scale - Thesis","Causal machine learning methods address causal-effect inference, causal reasoning, and causal structure discovery under uncertainty. This thesis studies statistical and structural uncertainty for scalable approaches working with large datasets and high-dimensional modalities including images, text, time series, and videos. Statistical uncertainty stems from fitting models to finite data, yielding plausible causal-effect ranges that contract with more training. Structural uncertainty reflects imprecise causal-structure knowledge requiring additional assumptions or interaction with the data-generating process.","Structural and Statistical Uncertainty in Observational Causal Machine Learning at Scale  \nAndrew Jesson  \nLinacre College University of Oxford  \nA thesis submitted for the degree of Doctor of Philosophy  \nMichaelmas 2023  \nIn loving memory of Mat Neuman  \nAcknowledgements  \nI would like to express my gratitude for the incredible support I have received from my family, Kathy and Don, Brad, Henry, Jen, and Mat.  \nI cannot overlook the contributions of those who have supported me from the beginning of a pivot that started 13 years ago in Montreal: Wilson, Tara, Phil, Laura, Marc, the Marosys, Laura M. , Chris, Meredith, Dalia, Monique, The Michaels, Gab, Alex, Blaˇz, Matina, Neda, Pi Cafe, the dedicated McGill Gym and Grounds teams, and Imagia Canexia Health.  \nI am grateful for the supportive community I have found at Oxford. In addition to everyone at OATML and Linacre College, I would particularly like to extend my thanks to Lisa Schut, Milad Alizadeh, Lewis Smith, Joost van Amersfoort, Luisa Zintgraf, Panagiotis Tigas, Mizu NishikawaToomey, Tim Rudner, Jannik Kossen, Muhammed Razzak, Lars Holdijk, Lorenz Kuhn, Kelsey Doerksen, and Eveline Lupien for their friendship and support.  \nI would also like to extend my appreciation to all collaborators: S¨oren Mindermann, Panagiotis Tigas, Alyson Douglas, Peter Manshausen, Joost Van Amersfoort, Andreas Kirsch, Lewis Smith, Oscar Key, Sebastian Farquhar, Parmida Atighehchian, Frederic Branchaud-Charron, Duncan Watson-Parris, Arash Mehrjou, Ashkan Soleymani, Pascal Notin, Stefan Bauer, Patrick Schwab, Yashas Annadani, Bernhard Sch¨olkopf, Nicolai Meinshausen, Ma¨elys Solal, Philip Stier, Desi Ivanova, Adam Foster, Clare Lyle, Miruna Oprescu, Marah Ghoummaid, Jacob Dorn, Nathan Kallus, Myrl Marmarelis, Elizabeth Haddad, Neda Jahanshad, Aram Galstyan, Greg Ver Steeg, Shreshth Malik, Salem Lahlou, Moksh Jain, Nikolay Malkin, Tristan Deleu, Yoshua Bengio, Chris Lu, Angelos Filos, Gunshi Gupta, and Jakob Foerster.  \nLastly, I would like to offer special thanks to Uri Shalit and my supervisor, Yarin Gal.  \nAbstract  \nCausal machine learning (Causal ML) tackles various tasks, including causal-effect inference, causal reasoning, and causal structure discovery. This thesis explores uncertainty for Causal ML methods that scale to large datasets and complex, high-dimensional input/output modalities, such as images, text, time series, and videos. Scalability is essential for efficiently processing vast amounts of information and predicting complex relationships.  \nAs we scale and achieve greater modeling flexibility, communicating the unknown becomes increasingly important. We examine two primary types of uncertainty: statistical and structural. Statistical uncertainty arises when fitting machine learning models to finite datasets. Addressing this uncertainty allows predicting a range of plausible causal effects that shrink with more training examples, facilitating better-informed decision-making and indicating areas needing improved understanding. Structural uncertainty arises from imprecise knowledge of the causal structure and generally requires further assumptions about the data-generating process or interaction with the world.  \nIn this thesis, we develop scalable Causal ML methods that navigate statistical and structural uncertainty effectively. We demonstrate the importance of considering scalability and uncertainty in Causal ML algorithm design and application, enhancing decision making and knowledge acquisition. Our contributions aim to advance the Causal machine learning field and provide a foundation for future research.  \nContents  \n1 Why do Scalability and Uncertainty Matter in Causal Machine Learning? 1  \n1.1 On Structural and Statistical Uncertainty ............... 2  \n1.2 Objective and Aims ............................ 4  \n1.3 Thesis Structure .............................. 4  \n2 Causal-effect Inference from Observational Data 7  \n2.1 Observational Data ...........................","cbCaij1YA2YGRjYt","https://ap.wps.com/l/cbCaij1YA2YGRjYt","pdf",10695707,1,140,"English","en",105,"# Why do Scalability and Uncertainty Matter in Causal Machine Learning?\n## On Structural and Statistical Uncertainty\n## Objective and Aims\n## Thesis Structure\n# Causal-effect Inference from Observational Data\n## Observational Data\n## Causal Estimands\n## Identifiability Conditions\n## Statistical Causal-Effect Estimands\n## Machine Learning for Causal-Effect Estimation\n## Uncertainty for Causal-Effect Inference\n# Methodological Background\n## Scalable Causal-Effect Inference\n## Scalable Statistical Uncertainty Quantification\n## Sensitivity Analysis for Structural Uncertainty\n## Active Learning and Experimental Design\n# Scalable Statistical Uncertainty for Causal Machine Learning\n## Statistical Uncertainty Estimands and Estimators","[{\"question\":\"What types of uncertainty are examined in this thesis for causal machine learning?\",\"answer\":\"The thesis focuses on statistical uncertainty, arising from fitting models to finite datasets, and structural uncertainty, caused by imprecise knowledge of the causal structure requiring additional assumptions about how data are generated.\"},{\"question\":\"How does scalability affect causal ML in the thesis?\",\"answer\":\"Scalability is treated as essential for efficiently processing vast amounts of information and for modeling complex high-dimensional modalities such as images, text, time series, and videos.\"},{\"question\":\"What is the practical goal of addressing statistical and structural uncertainty?\",\"answer\":\"The methods aim to communicate unknowns through uncertainty-aware predictions, improve decision-making, and highlight where further understanding of causal relationships is needed.\"}]","Structural and Statistical Uncertainty in Observational Causal Machine Learning at Scale - Thesis | PDF",1785674113,353,{"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},"structural-and-statistical-uncertainty-in-observational-causal-machine-learning-at-scale-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/structural-and-statistical-uncertainty-in-observational-causal-machine-learning-at-scale-thesis/117149/",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},"What types of uncertainty are examined in this thesis for causal machine learning?","Question",{"text":75,"@type":76},"The thesis focuses on statistical uncertainty, arising from fitting models to finite datasets, and structural uncertainty, caused by imprecise knowledge of the causal structure requiring additional assumptions about how data are generated.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does scalability affect causal ML in the thesis?",{"text":80,"@type":76},"Scalability is treated as essential for efficiently processing vast amounts of information and for modeling complex high-dimensional modalities such as images, text, time series, and videos.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the practical goal of addressing statistical and structural uncertainty?",{"text":84,"@type":76},"The methods aim to communicate unknowns through uncertainty-aware predictions, improve decision-making, and highlight where further understanding of causal relationships is needed.","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"]