[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-128241-en":3,"doc-seo-128241-105":31,"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},128241,2336475104736,"Quinn","https://ap-avatar.wpscdn.com/avatar/22000c4c5e0e5b17e70?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786591360781797222",8,"Research & Report","Continuous Management of Machine Learning-Based Application Behavior - Paper abstract and approach summary","Modern ML-driven applications exhibit non-deterministic behavior that can affect the whole lifecycle from design through operation. The work addresses the need to continuously guarantee stable non-functional behavior despite evolving models and contextual changes. It emphasizes monitoring, verification, and maintenance of non-functional properties such as privacy, confidentiality, fairness, and explainability. A multi-model, ML-agnostic two-step architecture is proposed: model assessment validates properties at development time and model substitution ensures continuous stable support during operation. Experiments target fairness in a real-world scenario.","This article has been accepted for publication in IEEE Transactions on Services Computing. This is the author's version which has not been fully edited and content may change prior to final publication. Citation information: DOI 10. 1109/TSC.2024.3486226  \n1  \nContinuous Management of Machine Learning-Based Application Behavior  \nMarco Anisetti Senior Member, IEEE, Claudio A. Ardagna Senior Member, IEEE, Nicola Bena Student Member, IEEE, Ernesto Damiani Senior Member, IEEE, Paolo G. Panero  \nAbstract—Modern applications are increasingly driven by Machine Learning (ML) models whose non-deterministic behavior is affecting the entire application life cycle from design to operation. The pervasive adoption of ML is urgently calling for approaches that guarantee a stable non-functional behavior of ML-based applications over time and across model changes. To this aim, non-functional properties of ML models, such as privacy, confidentiality, fairness, and explainability, must be monitored, verified, and maintained. Existing approaches mostly focus on i) implementing solutions for classifier selection according to the functional behavior of ML models, ii) finding new algorithmic solutions, such as continuous re-training. In this paper, we propose a multi-model approach that aims to guarantee a stable non-functional behavior of ML-based applications. An architectural and methodological approach is provided to compare multiple ML models showing similar non-functional properties and select the model supporting stable  \nnon-functional behavior over time according to (dynamic and unpredictable) contextual changes. Our approach goes beyond the state of the art by providing a solution that continuously guarantees a stable non-functional behavior of ML-based applications, is ML algorithm-agnostic, and is driven by non-functional properties assessed on the ML models themselves. It consists of a two-step process working during application operation, where model assessment verifies non-functional properties of ML models trained and selected at development time, and model substitution guarantees continuous and stable support of non-functional properties. We experimentally evaluate our solution in a real-world scenario focusing on non-functional property fairness.  \nIndex Terms—Assurance, Machine Learning, Multi-Armed Bandit, Non-Functional Properties  \n~~ ~~ ✦ ~~ ~~  \n1 INTRODUCTION  \nMachine Learning (ML) has become the technique of choice to provide advanced functionalities and carry out tasks hardly achievable by traditional control and optimization algorithms [1] . Even the behavior, orchestration, and deployment parameters of distributed systems and services, possibly offered on the cloud-edge continuum, are increasingly based on ML models [2] . Concerns about the black-box nature of ML have led to a societal push that involves all components of society (policymakers, regulators, academic and industrial stakeholders, citizens) towards trustworthy and transparent ML, giving rise to legislative initiatives on artificial intelligence (e.g., the AI Act in Europe [3]) .  \nThis scenario introduces the need for solutions that continuously guarantee a stable non-functional behavior of ML-based applications, a task that is significantly more complex than mere QoS-based selection and composition (e.g., [4], [5], [6]) . The focus of such a task is to assess thenon-functional properties of ML models, such as privacy, confidentiality, fairness, and explainability, over time and across changes. The non-functional assessment of ML-based applications behavior has to cope with the ML models’complexity, low transparency, and continuous evolution [7],[8] . ML models in fact are affected by model and data  \n• M. Anisetti, C. A. Ardagna, N. Bena, E. Damiani, Paolo G. Panero are with the Department of Computer Science, Universit`a degli Studidi Milano, Milano, Italy. E. Damiani is also with Khalifa University, Abu Dhabi, UAE.  \nE-mail: {firstname.lastname}@unimi.i","cbCaihgwUknhdan7","https://ap.wps.com/l/cbCaihgwUknhdan7","pdf",1164965,2,1,14,"English","en",105,"# Introduction\n## Problem and motivation\n## Related work overview\n## Proposed multi-model approach\n## Experimental evaluation","[{\"question\":\"Why is continuous management needed for ML-based application behavior?\",\"answer\":\"ML models are non-deterministic and evolve over time, making non-functional behavior unstable across model changes and contextual shifts. The paper argues that stable non-functional assurance is more complex than QoS-based selection.\"},{\"question\":\"Which non-functional properties are the focus of the proposed approach?\",\"answer\":\"The approach targets privacy, confidentiality, fairness, and explainability, treating them as properties to be monitored, verified, and maintained over time.\"},{\"question\":\"How does the two-step process work during application operation?\",\"answer\":\"Model assessment verifies non-functional properties of ML models trained and selected at development time, while model substitution guarantees continuous and stable support by replacing models when needed during operation.\"}]","Continuous Management of Machine Learning-Based Application Behavior - Paper abstract and approach summary | PDF",1785946040,35,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":29},"continuous-management-of-machine-learning-based-application-behavior-paper-abstract-and-approach-summary","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,48,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":20},"https://docshare.wps.com/document/","Document",{"item":49,"name":12,"@type":44,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/continuous-management-of-machine-learning-based-application-behavior-paper-abstract-and-approach-summary/128241/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"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},"Why is continuous management needed for ML-based application behavior?","Question",{"text":76,"@type":77},"ML models are non-deterministic and evolve over time, making non-functional behavior unstable across model changes and contextual shifts. The paper argues that stable non-functional assurance is more complex than QoS-based selection.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Which non-functional properties are the focus of the proposed approach?",{"text":81,"@type":77},"The approach targets privacy, confidentiality, fairness, and explainability, treating them as properties to be monitored, verified, and maintained over time.",{"name":83,"@type":74,"acceptedAnswer":84},"How does the two-step process work during application operation?",{"text":85,"@type":77},"Model assessment verifies non-functional properties of ML models trained and selected at development time, while model substitution guarantees continuous and stable support by replacing models when needed during operation.","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":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]