[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117787-en":3,"doc-seo-117787-105":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},117787,687197100911,"Himbo","https://ap-avatar.wpscdn.com/avatar/a000239b6f1da00475?x-image-process=image/resize,m_fixed,w_180,h_180&k=1785132997149421697",8,"Research & Report","Causal Inference Methods for Supporting, Understanding, and Improving Decision-Making","Causality and the ability to reason about cause-and-effect relationships are central to decision-making. This thesis develops causal inference approaches using machine learning to support, understand, and improve decisions, with a focus on healthcare. It proposes methods to estimate causal effects of interventions from observational data, including counterfactual outcomes for continuous-valued treatments using generative adversarial networks. It also addresses temporal confounding via a sequence-to-sequence model and accounts for multi-cause hidden confounders. Further, it integrates counterfactual reasoning into batch inverse reinforcement learning to model expert preferences over counterfactual outcomes, enabling interpretability and clearer trade-offs in decision behavior.","Causal Inference Methods for Supporting, Understanding, and Improving Decision-Making  \nIoana Bica Wolfson College  \nUniversity of Oxford  \nA thesis submitted for the degree of Doctor of Philosophy  \nHilary 2022  \nAcknowledgements  \nPersonal  \nThis thesis would not have been possible without the help and support of my advisor, my colleagues, my family, and my friends to which I am extremely grateful!  \nFirst and foremost, I would like to thank my advisor, Professor Mihaela van der Schaar for her guidance throughout my DPhil and for pushing me to achieve my best. Thankyou for encouraging me to go beyond my comfort zone, to think deeply about research problems, to be creative, and to think big and outside the box! I am also grateful foryour research vision which has significantly shaped this thesis and for always providing me with new ideas, insights, and invaluable advice which have had a profound impact on my development as a researcher.  \nI would also like to thank the members of the van der Schaar Lab for the countless insightful research discussions and engaging group meetings from which I have learned so much. To begin with, I would like to thank Ahmed M. Alaa for his invaluable support and guidance throughout the first year of my DPhil. Moreover, I am also grateful to my close collaborators James Jordon and Dan Jarrett. It has been a pleasure to brainstorm research ideas with you, learn from you and, work together on several research papers! I would also like to thank my other collaborators and co-authors from the lab: Trent Kyono, Alihan Hüyük, Jeroen Berrevoets, Alicia Curth, Jinsung Yoon, Yao Zhang, Zhaozhi Qian, Alex Chan, Changhee Lee, Evgeny Saveliev, Jonathan Crabbé, Nabeel Seedat, Fergus Imrie and Kamile˙ Stankevii¯ute˙ . Being able to discuss ideas and work with all of you has been one of my favorite parts of the DPhil and it has been a privilege to have such talented people as my colleagues.  \nI am also deeply grateful to Professor Pietro Lió for advising my undergraduate and master’s research projects which motivated me to pursue a DPhil. Thank you for your unwavering support, advice and for always ensuring I was doing well!  \nFurthermore, I would like to thank my advisers and collaborators from DeepMind who provided me with a wonderful research internship experience from which I have learned a lot and which has helped me grow as a researcher. Thank you, Jovana Mitrovi, for being an amazing and supportive host and for your invaluable advice and guidance which has contributed greatly to the success of my internship project. I am also extremely grateful to Nenad Tomašev, Brian McWilliams, Rzvan Pa¸scanu, Lars Buesing, and Charles Blundell for their helpful advice throughout my internship.  \nMoreover, I would like to thank Professor Pawan Mudigonda and Professor Victor Adrian Prisacariu for their support in helping me to continue my DPhil at the University of Oxford.  \nIn addition, I am grateful to my DPhil viva examiners, Professor Alison Noble and Dr Danielle Belgrave for the insightful questions and conversations during my viva examination and for their suggestions for future work.  \nI am also grateful to my friend Malavika Nair, who has been by my side throughout my academic journey of becoming a researcher since my undergraduate studies and who has provided me with invaluable advice and support. In addition, I would also like to thank my friend Bianca Danciu for always being there for me and for always believing in me.  \nLast but not least, I am eternally grateful to my parents, Ion and Elena Bica, for their unconditional love, support, and encouragement. This thesis would not have been possible without all of the opportunities they have provided me with throughout my life. I am also deeply grateful to my little sister, Ana-Maria Bica, for her love and support, for taking care of me, for putting up with me, and for always being there for me.  \nInstitutional  \nI am incredibly grateful to the Alan Turing Instit","cbCairlHIVpW2sg2","https://ap.wps.com/l/cbCairlHIVpW2sg2","pdf",13334797,1,259,"English","en",105,"# Abstract\n## Supporting decision-making with causal effect estimation\n## Counterfactual outcomes for continuous-valued interventions\n## Temporal causal inference under time-dependent confounding\n## Handling multi-cause hidden confounders in temporal data\n## Interpretable expert modeling via batch inverse reinforcement learning","[{\"question\":\"What is the main focus of the thesis?\",\"answer\":\"The thesis focuses on causal inference methods that use machine learning to support, understand, and improve decision-making, with emphasis on healthcare applications.\"},{\"question\":\"How does the thesis estimate treatment effects from observational data?\",\"answer\":\"It introduces causal inference tools that estimate causal effects of interventions from observational data such as electronic health records.\"},{\"question\":\"What approaches are proposed for counterfactual and temporal problems?\",\"answer\":\"It proposes a generative adversarial network method for counterfactual outcomes with continuous-valued interventions, and a sequence-to-sequence model with domain adversarial training to address time-dependent confounding.\"}]","Causal Inference Methods for Supporting, Understanding, and Improving Decision-Making | PDF",1785679560,653,{"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},"causal-inference-methods-for-supporting-understanding-and-improving-decision-making","",{"@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/causal-inference-methods-for-supporting-understanding-and-improving-decision-making/117787/",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 is the main focus of the thesis?","Question",{"text":75,"@type":76},"The thesis focuses on causal inference methods that use machine learning to support, understand, and improve decision-making, with emphasis on healthcare applications.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the thesis estimate treatment effects from observational data?",{"text":80,"@type":76},"It introduces causal inference tools that estimate causal effects of interventions from observational data such as electronic health records.",{"name":82,"@type":73,"acceptedAnswer":83},"What approaches are proposed for counterfactual and temporal problems?",{"text":84,"@type":76},"It proposes a generative adversarial network method for counterfactual outcomes with continuous-valued interventions, and a sequence-to-sequence model with domain adversarial training to address time-dependent confounding.","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"]