[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125568-en":3,"doc-seo-125568-105":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},125568,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","Addressing uncertainty in the safety assurance of machine-learning","Addressing uncertainty in the safety assurance of machine-learning examines why proving safety for machine-learning in safety-critical cyber-physical systems remains a major obstacle to deployment. The work analyzes how uncertainty shapes confidence in safety assurance arguments and links it to both the complexity of learning models and the complexity of the target tasks. Using uncertainty definitions and an exemplary assurance-argument structure, it identifies common argument weaknesses through systematic review of asserted context, evidence, and inference, and derives structural requirements supported by evidence. The study concludes that robust safety claims require qualitative arguments supported by quantitative evidence, continuously refined to reduce residual and emerging uncertainties after deployment.","TYPE Hypothesis and Theory PUBLISHED 06 April 2023  \nDOI 10. 3389/fcomp.2023.1132580  \nOPEN ACCESS  \nEDITED BY  \nXiaowei Huang,  \nUniversity of Liverpool, United Kingdom  \nREVIEWED BY  \nJianwen Li,  \nEast China Normal University, China Panagiotis Katsaros,  \nAristotle University of Thessaloniki, Greece  \n*CORRESPONDENCE  \nSimon Burton  \n [simon.burton@iks.fraunhofer.de](simon.burton@iks.fraunhofer.de)  \nSPECIALTY SECTION  \nThis article was submitted to Software,  \na section of the journal Frontiers in Computer Science  \nRECEIVED 27 December 2022  \nACCEPTED 20 March 2023  \nPUBLISHED 06 April 2023  \nCITATION  \nBurton S and Herd B (2023) Addressing uncertainty in the safety assurance of machine-learning.  \nFront. Comput. Sci. 5:1132580 .  \ndoi: 10.3389/fcomp.2023.1132580  \nCOPYRIGHT  \n© 2023 Burton and Herd. This is an  \nopen-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.  \nAddressing uncertainty in the safety assurance of machine-learning  \nSimon Burton* and Benjamin Herd  \nFraunhofer Institute for Cognitive Systems, Munich, Germany  \nThere is increasing interest in the application of machine learning (ML) technologies to safety-critical cyber-physical systems, with the promise of increased levels of autonomy due to their potential for solving complex perception and planning tasks. However, demonstrating the safety of ML is seen as one of the most challenging hurdles to their widespread deployment for such applications. In this paper we explore the factors which make the safety assurance of ML such a challenging task. In particular we address the impact of uncertainty on the confidence in ML safety assurance arguments. We show how this uncertainty is related to complexity in the ML models as well as the inherent complexity of the tasks that they are designed to implement. Based on definitions of uncertainty as well as an exemplary assurance argument structure, we examine typical weaknesses in the argument and how these can be addressed. The analysis combines an understanding of causes of insu􀀈ciencies in ML models with a systematic analysis of the types of asserted context, asserted evidence and asserted inference within the assurance argument. This leads to a systematic identification of requirements on the assurance argument structure as well as supporting evidence. We conclude that a combination of qualitative arguments combined with quantitative evidence are required to build a robust argument for safety-related properties of ML functions that is continuously refined to reduce residual and emerging uncertainties in the arguments after the function has been deployed into the target environment.  \nKEYWORDS  \nmachine learning, safety, assurance arguments, cyber-physical systems, uncertainty, complexity  \n1. Introduction  \nRecent advances in the 􀀂eld of arti􀀂cial intelligence (AI), and in particular Machine Learning (ML), have led to increased interest in the application of ML to cyber-physical systems such as autonomous vehicles and industrial robotics. Such systems have the potential to increase safety through increased automation, for example by reducing the number of human-induced accidents, or allowing systems to operate in hazardous environments without direct human control. However, the malfunctioning of such systems can lead to severe harm to users, bystanders and the environment. There is therefore a clear need to demonstrate that safety-critical systems that utilize ML are acceptably safe. As a consequence, the 􀀂eld of trustworthy and safe AI is also receiving attention from a regulatory and standards perspectives. Examples of which are the EU","cbCailo2Tk906d3j","https://ap.wps.com/l/cbCailo2Tk906d3j","pdf",1240786,1,17,"English","en",105,"# Introduction\n# Uncertainty and complexity in ML safety assurance\n# Assurance argument structure and weaknesses\n# Requirements and supporting evidence\n# Conclusion","[{\"question\":\"Why is safety assurance for machine-learning considered challenging?\",\"answer\":\"Because demonstrating safety for ML-based functions in safety-critical cyber-physical systems is difficult, especially when complex models and complex tasks produce specification and performance insufficiencies that drive assurance uncertainty.\"},{\"question\":\"How does uncertainty affect confidence in ML safety assurance arguments?\",\"answer\":\"Uncertainty impacts the confidence that safety assurance arguments provide, and it is connected to the complexity of ML models as well as the inherent complexity of the tasks the models implement.\"},{\"question\":\"What weaknesses are examined in typical safety assurance arguments?\",\"answer\":\"The paper examines common weaknesses in argument structure by analyzing asserted context, asserted evidence, and asserted inference, and explains how insufficiencies in ML models contribute to resulting assurance uncertainty.\"}]","Addressing uncertainty in the safety assurance of machine-learning | PDF",1785899916,43,{"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},"addressing-uncertainty-in-the-safety-assurance-of-machine-learning","",{"@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/addressing-uncertainty-in-the-safety-assurance-of-machine-learning/125568/",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","2026-08-05",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},"Why is safety assurance for machine-learning considered challenging?","Question",{"text":75,"@type":76},"Because demonstrating safety for ML-based functions in safety-critical cyber-physical systems is difficult, especially when complex models and complex tasks produce specification and performance insufficiencies that drive assurance uncertainty.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does uncertainty affect confidence in ML safety assurance arguments?",{"text":80,"@type":76},"Uncertainty impacts the confidence that safety assurance arguments provide, and it is connected to the complexity of ML models as well as the inherent complexity of the tasks the models implement.",{"name":82,"@type":73,"acceptedAnswer":83},"What weaknesses are examined in typical safety assurance arguments?",{"text":84,"@type":76},"The paper examines common weaknesses in argument structure by analyzing asserted context, asserted evidence, and asserted inference, and explains how insufficiencies in ML models contribute to resulting assurance uncertainty.","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"]