[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-128572-en":3,"doc-seo-128572-105":30,"detail-sidebar-cat-0-en-105":92},{"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},128572,549768064778,"Finn","https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8",8,"Research & Report","Bayesian Optimization in Adverse Scenarios - Doctor of Philosophy Thesis","Optimization problems with expensive-to-evaluate objective functions are common across scientific and industrial workflows. Bayesian optimization is widely used because it offers theoretical performance guarantees and strong sample efficiency for expensive, often black-box, functions. This thesis addresses four practical adverse scenarios where standard methods underperform. It develops approaches for robust design under input noise, sample-efficient high-dimensional multiobjective optimization for complex design models, and Bayes-optimal non-myopic acquisition for variable-cost and multi-fidelity settings, including hardware-aware neural architecture search. It also proposes a probabilistic reparameterization technique to enable effective policies when the search space is discrete or mixed, providing a practical playbook for challenging conditions.","Bayesian Optimization in Adverse Scenarios  \nSamuel James Daulton  \nSt. Catherine’s College  \nUniversity of Oxford  \nA thesis submitted for the degree of Doctor of Philosophy  \nHilary 2023  \nAbstract  \nOptimization problems with expensive-to-evaluate objective functions are ubiquitous in scientific and industrial settings. Bayesian optimization has gained widespread acclaim for optimizing expensive (and often black box) functions due to its theoretical performance guarantees and empirical sample efficiency in a variety of settings. Nevertheless, many practical scenarios remain where prevailing Bayesian optimization techniques fall short. We consider four such scenarios. First, we formalize the optimization problem where the goal is to identify robust designs with respect to multiple objective functions that are subject to input noise. Such robust design problems frequently arise, for example, in manufacturing settings where fabrication can only be performed with limited precision. We propose a method that identifies a set of optimal robust designs, where each design provides probabilistic guarantees jointly on multiple objectives. Second, we consider sample-efficient high-dimensional multiobjective optimization. This line of research is motivated by the challenging task of designing optical displays for augmented reality to optimize visual quality and efficiency, where the designs are specified by high-dimensional parameterizations governing complex geometries. Our proposed trust-region based algorithm yields order-of-magnitude improvements in sample complexity on this problem. Third, we consider multi-objective optimization of expensive functions with variable-cost, decoupled, and/or multi-fidelity evaluations and propose a Bayes-optimal, non-myopic acquisition function, which significantly improves sample efficiency in scenarios with incomplete information. We apply this to hardware-aware neural architecture search where the objective, on-device latency and model accuracy, can often be evaluated independently. Fourth, we consider the setting where the search space consists of discrete (and potentially continuous) parameters. We propose a theoretically grounded technique that uses a probabilistic reparameterization to transform the discrete or mixed inner optimization problem into a continuous one leading to more effective Bayesian optimization policies. Together, this thesis provides a playbook for Bayesian optimization in several practical adverse scenarios.  \nBayesian Optimization in Adverse  \nScenarios  \nSamuel James Daulton St. Catherine’s College  \nUniversity of Oxford  \nA thesis submitted for the degree of Doctor of Philosophy  \nHilary 2023  \nThis thesis is dedicated to Sandra Lynn Daulton for her unwavering affirmation in all pursuits of education.  \nAcknowledgements  \nPersonal  \nThank you to my advisor Mike Osborne for your support throughout my DPhil. I would also like to thank Max Balandat and Eytan Bakshy for your advice, mentorship and support over the last 5 years.  \nThanks to my parents Sandra Daulton and James Daulton; my step-parents Roz Ho and Michael Zeidman; my sister Melanie Daulton; and all of my family and friends.  \nAnd thanks to my partner Sara Violett for tolerating late nights and Sunday workdays—at the expense of adventures and contributing to good housekeeping.  \nInstitutional  \nI am grateful for the support of Meta and the University of Oxford during my DPhil.  \nAbstract  \nOptimization problems with expensive-to-evaluate objective functions are ubiquitous in scientific and industrial settings. Bayesian optimization has gained widespread acclaim for optimizing expensive (and often black box) functions due to its theoretical performance guarantees and empirical sample efficiency in a variety of settings. Nevertheless, many practical scenarios remain where prevailing Bayesian optimization techniques fall short. We consider four such scenarios. First, we formalize the optimization problem where the g","cbCaigttuiDUj5tc","https://ap.wps.com/l/cbCaigttuiDUj5tc","pdf",12628372,1,269,"English","en",105,"# Abstract\n## Robust design under input noise\n## Sample-efficient high-dimensional multiobjective optimization\n## Variable-cost and multi-fidelity multiobjective acquisition\n## Discrete or mixed parameter spaces","[{\"question\":\"What adverse scenarios does the thesis address in Bayesian optimization?\",\"answer\":\"It considers four scenarios where prevailing Bayesian optimization techniques fall short: robust design with input noise, sample-efficient high-dimensional multiobjective optimization, variable-cost decoupled and/or multi-fidelity evaluations, and optimization over discrete (or mixed) parameter spaces.\"},{\"question\":\"How does the work handle robust designs with multiple noisy objective functions?\",\"answer\":\"It formalizes robust optimization with input noise and proposes a method that identifies a set of optimal robust designs, giving probabilistic guarantees jointly across multiple objectives.\"},{\"question\":\"What acquisition and transformation ideas are proposed to improve efficiency with incomplete information and discrete parameters?\",\"answer\":\"For variable-cost and multi-fidelity settings, it proposes a Bayes-optimal, non-myopic acquisition function to improve sample efficiency under incomplete information. For discrete or mixed search spaces, it uses probabilistic reparameterization to convert the inner optimization into a continuous problem for more effective Bayesian optimization policies.\"}]","Bayesian Optimization in Adverse Scenarios - Doctor of Philosophy Thesis | PDF",1786001815,678,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"bayesian-optimization-in-adverse-scenarios-doctor-of-philosophy-thesis","",{"@graph":36,"@context":86},[37,54,69],{"@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/bayesian-optimization-in-adverse-scenarios-doctor-of-philosophy-thesis/128572/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-24","2026-08-06",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What adverse scenarios does the thesis address in Bayesian optimization?","Question",{"text":76,"@type":77},"It considers four scenarios where prevailing Bayesian optimization techniques fall short: robust design with input noise, sample-efficient high-dimensional multiobjective optimization, variable-cost decoupled and/or multi-fidelity evaluations, and optimization over discrete (or mixed) parameter spaces.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the work handle robust designs with multiple noisy objective functions?",{"text":81,"@type":77},"It formalizes robust optimization with input noise and proposes a method that identifies a set of optimal robust designs, giving probabilistic guarantees jointly across multiple objectives.",{"name":83,"@type":74,"acceptedAnswer":84},"What acquisition and transformation ideas are proposed to improve efficiency with incomplete information and discrete parameters?",{"text":85,"@type":77},"For variable-cost and multi-fidelity settings, it proposes a Bayes-optimal, non-myopic acquisition function to improve sample efficiency under incomplete information. For discrete or mixed search spaces, it uses probabilistic reparameterization to convert the inner optimization into a continuous problem for more effective Bayesian optimization policies.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]