[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-128343-en":3,"doc-seo-128343-105":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},128343,962085564549,"Genevieve","https://ap-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","Statistical Methods leveraging Uncertainties in Machine Learning - Dissertation","In data-driven applications, machine learning models face safety, reliability, and interpretability requirements. Hidden factors and limited data availability can degrade performance and increase predictive uncertainty. This dissertation centers on recognizing, quantifying, and reducing key uncertainty types—aleatoric and epistemic—especially in domains such as autonomous driving, medical diagnostics, finance, weather forecasting, and industrial production. Adoption is often slowed by computational cost and model complexity, motivating probabilistic strategies plus more efficient approaches that reduce complexity while preserving robustness. The work presents methods across four publications, including uncertainty-aware preprocessing, probabilistic interpolation, and expert knowledge integration.","Statistical Methods leveraging Uncertainties  \nin Machine Learning  \nDissertation  \nan der Fakultät für Mathematik, Informatik und Statistik der Ludwig-Maximilians-Universität München  \neingereicht von Stefan Michael Stroka am 12.06.2025  \nStatistical Methods leveraging Uncertainties in Machine Learning  \nErstgutachter: Prof. Dr . Christian Heumann,  \nInstitut für Statistik, LMU München  \nZweitgutachter: Prof. Dr . Volker Schmid,  \nInstitut für Statistik, LMU München  \nDrittgutachter: Prof. Dr . Martin Spieß ,  \nInstitut für Psychologie, Universität Hamburg  \nVorgelegt von: Stefan Stroka  \nTag der mündlichen Prüfung: 14 . November 2025  \nDanksagung  \nZu Beginn möchte ich mich herzlichst bei all jenen bedanken, die mich bei meiner persönlichen und fachlichen Weiterentwicklung gefördert und unterstützt haben. Ohne diese Unterstützung wäre die vorliegende Arbeit in dieser Form nicht möglich gewesen.  \nAn erster Stelle möchte ich mich ausdrücklich bei meinem Doktorvater, Prof. Dr. Christian Heumann, bedanken, der sich die Zeit genommen hat, mit mir die komplexen Aufgabenstellungen zu erarbeiten und mich während der gesamten Arbeit stets mit wertvollen Anregungen, seiner fachlichen Expertise und geduldiger Unterstützung begleitet hat.  \nMein besonderer Dank gilt meinem Vater Wolfgang Stroka, meinem Bruder Christian Stroka und Janina Wanzke, die mir durch ihre unermüdliche Unterstützung, ihren Zuspruch und ihren festen Glauben an mich maßgeblich ermöglicht haben, die Promotion anzustreben und schließlich diese Arbeit zu vollenden. Darüber hinaus danke ich auch meiner gesamten Familie und meinen Freunden, die mir stets Rückhalt gegeben und mich ermutigt haben. Ihre Geduld, ihr Verständnis und ihre fortwährende Unterstützung haben mir geholfen, auch herausfordernde Zeiten zu meistern.  \nMein Dank gilt auch meinen Kollegen bei ams Osram, die mich durch ihre Zusammenarbeit und ihr offenes Ohr stets unterstützt haben. Ihr Feedback und das gemeinsame Arbeiten haben mir in vielen Phasen der Promotion sehr geholfen.  \nAbschließend danke ich allen, die auf unterschiedliche Weise zum Gelingen dieser Arbeit beigetragen haben.  \nAbstract  \nIn today’s data-driven landscape, machine learning methods are increasingly applied in domains that demand high levels of safety, reliability, and interpretability. However, hidden influencing factors and limited data availability can significantly impair model performance and amplify predictive uncertainty. Recognizing, quantifying, and—where possible—reducing uncertainties such as aleatoric and epistemic uncertainty has therefore become a central concern. This is particularly true in critical fields like autonomous driving, medical diagnostics, finance, weather forecasting, and industrial production, where dependable predictions are not merely advantageous, but essential. Despite its relevance, the broader adoption of uncertainty quantification in practice is often hindered by high computational demands and growing model complexity. Furthermore, aleatoric uncertainty—stemming from noise and imperfections in the data itself—poses a fundamental challenge to the reliability of data-driven models. This dissertation explores multiple strategies for uncertainty quantification across four publications. These contributions examine both the necessity and the practical implementation of probabilistic techniques, while also introducing novel, less computationally intensive methods that reduce model complexity without sacrificing robustness.  \nPublication 1:  \nThe first study addresses epistemic and aleatoric uncertainties in the preprocessing phase of industrial production modeling. Aleatoric uncertainties are predefined based on expert experience, setting the bounds for acceptable input variation. Given the limited spatial distribution of measurement data, uncertainty-aware interpolation is applied for data augmentation. Probabilistic Gaussian Process Regression is employed to model prediction intervals and serve as ","cbCainKx3OXeml2t","https://ap.wps.com/l/cbCainKx3OXeml2t","pdf",11619958,1,134,"English","en",105,"# Danksagung\n# Abstract\n# Publication 1\n## Uncertainty-aware preprocessing and modeling\n# Publication 2\n## Regression-classification with fuzzy logic\n# Publication 3\n## Expert-derived knowledge for robustness\n# Publication 4\n## Probabilistic training with small datasets\n# Zusammenfassung","[{\"question\":\"What problem does the dissertation address in machine learning?\",\"answer\":\"It addresses how hidden influences and limited data can harm predictive performance and increase uncertainty, making reliable predictions difficult in safety- and mission-critical domains.\"},{\"question\":\"Which types of uncertainty are studied?\",\"answer\":\"The dissertation focuses on aleatoric uncertainty (noise and data imperfections) and epistemic uncertainty (uncertainty due to limited knowledge or missing information).\"},{\"question\":\"How is uncertainty quantified and reduced across the four publications?\",\"answer\":\"Across the studies, it uses uncertainty-aware preprocessing and probabilistic techniques such as Gaussian Process Regression, combines regression and ordinal classification with a customized loss, incorporates expert-derived knowledge without requiring experts at inference time, and applies probabilistic approaches for training with small datasets.\"}]","Statistical Methods leveraging Uncertainties in Machine Learning - Dissertation | PDF",1785946956,338,{"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},"statistical-methods-leveraging-uncertainties-in-machine-learning-dissertation","",{"@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/statistical-methods-leveraging-uncertainties-in-machine-learning-dissertation/128343/",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","2026-08-23","2026-08-05",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 problem does the dissertation address in machine learning?","Question",{"text":76,"@type":77},"It addresses how hidden influences and limited data can harm predictive performance and increase uncertainty, making reliable predictions difficult in safety- and mission-critical domains.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Which types of uncertainty are studied?",{"text":81,"@type":77},"The dissertation focuses on aleatoric uncertainty (noise and data imperfections) and epistemic uncertainty (uncertainty due to limited knowledge or missing information).",{"name":83,"@type":74,"acceptedAnswer":84},"How is uncertainty quantified and reduced across the four publications?",{"text":85,"@type":77},"Across the studies, it uses uncertainty-aware preprocessing and probabilistic techniques such as Gaussian Process Regression, combines regression and ordinal classification with a customized loss, incorporates expert-derived knowledge without requiring experts at inference time, and applies probabilistic approaches for training with small datasets.","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"]