[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117773-en":3,"doc-seo-117773-105":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},117773,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","Unexpectedly Useful - Convergence Bounds And Real-World Distributed Learning","Convergence bounds help estimate performance of distributed machine learning tasks before execution, yet their real predictive and improvement value is often unclear. This work runs experiments to evaluate how such bounds forecast and enhance real-world federated learning. Results show bounds are typically loose and their relative scale tracks training rather than testing loss. Notably, specific bound quantities can identify clients most likely to contribute without disclosing dataset quality or size.","POLITECNICO DI TORINO  \nRepository ISTITUZIONALE  \nUnexpectedly Useful: Convergence Bounds And Real-World Distributed Learning  \nOriginal  \nUnexpectedly Useful: Convergence Bounds And Real-World Distributed Learning / Malandrino, Francesco; Chiasserini, Carla Fabiana. -STAMPA. - (2023), pp. 76-79. (Intervento presentato al convegno 2023 15th International Conference on Machine Learning and Computing (ICMLC 2023) tenutosi a Zhuhai (China) nel Feb. 2023)[10 . 1145/3587716 .3587728] .  \nAvailability:  \nThis version is available at: 11583/2973549 since: 2022-12-01T14:12:17Z  \nPublisher: ACM  \nPublished  \nDOI:10.1145/3587716.3587728  \nTerms of use:  \nThis article is made available under terms and conditions as specified in the corresponding bibliographic description in the repository  \nPublisher copyright  \nACM postprint/Author's Accepted Manuscript  \n(Article begins on next page)  \n19 October 2023  \nUnexpectedly Useful: Convergence Bounds And Real-World Distributed Learning  \nFrancesco Malandrino  \nCNR-IEIIT and CNIT Torino, Italy  \nCarla Fabiana Chiasserini  \nPolitecnico di Torino, CNR-IEIIT, and CNIT Torino, Italy  \nABSTRACT  \nConvergence bounds are one of the main tools to obtain information on the performance of a distributed machine learning task, before running the task itself. In this work, we perform a set of experiments to assess to which extent, and in which way, such bounds can predict and improve the performance of real-world distributed (namely, federated) learning tasks. We find that, as can be expected given the way they are obtained, bounds are quite loose and their relative magnitude reflects the training rather than the testing loss. More unexpectedly, we find that some of the quantities appearing in the bounds turn out to be very useful to identify the clients that are most likely to contribute to the learning process, without requiring the disclosure of any information about the quality or size of their datasets. This suggests that further research is warranted on the ways – often counter-intuitive – in which convergence bounds can be exploited to improve the performance of real-world distributed learning tasks.  \nACM Reference Format:  \nFrancesco Malandrino and Carla Fabiana Chiasserini. 2023. Unexpectedly Useful: Convergence Bounds And Real-World Distributed Learning. In Proceedings of International Conference on Machine Learning and Computing (ICMLC’23). ACM, New York, NY, USA, 4 pages. [https://doi.org/10.1145/](https://doi.org/10.1145/)[ ](https://doi.org/10.1145/)nnnnnnn.nnnnnnn  \n1 INTRODUCTION  \nIt would be hard to overstate the importance of machine learning (ML) for a growing number of aspects of technology and, indeed, of our daily lives. Furthermore, owing to the growing complexity of the learning tasks to perform, to the ever-increasing amount of resources they require, and to the need to keep data local, a lot of today’s learning is distributed, i.e., it requires the cooperation of multiple learning nodes, leveraging the help of a learning server.  \nA prominent example of distributed learning is represented by the Federated Learning (FL) paradigm, which operates [1] by performing five main steps, as summarized in Fig. 1:  \n(1) each learning node trains a local model, leveraging on-device data;  \n(2) after one or more local epochs, learning nodes send their current model to the server;  \nPermission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions [from permissions@acm.org](from permissions@acm.org).  \nICMLC’23, Februa","cbCaioZnDHklWSyl","https://ap.wps.com/l/cbCaioZnDHklWSyl","pdf",807200,1,5,"English","en",105,"# Abstract\n# Introduction\n## Federated learning paradigm and iteration steps\n## Factors affecting overall performance\n## Convergence bounds and why they matter","[{\"question\":\"What problem does the work address about convergence bounds?\",\"answer\":\"It tests to what extent convergence bounds can predict and improve performance in real-world distributed (federated) learning tasks.\"},{\"question\":\"What do the experiments find about how tight convergence bounds are?\",\"answer\":\"The bounds are generally quite loose, and their relative magnitude reflects training loss more than testing loss.\"},{\"question\":\"How can convergence-bound quantities help in federated learning without sharing data?\",\"answer\":\"Certain quantities in the bounds help identify clients most likely to contribute to learning while not requiring disclosure of dataset quality or size.\"}]","Unexpectedly Useful - Convergence Bounds And Real-World Distributed Learning | PDF",1785679480,13,{"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},"unexpectedly-useful-convergence-bounds-and-real-world-distributed-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/unexpectedly-useful-convergence-bounds-and-real-world-distributed-learning/117773/",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 problem does the work address about convergence bounds?","Question",{"text":75,"@type":76},"It tests to what extent convergence bounds can predict and improve performance in real-world distributed (federated) learning tasks.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What do the experiments find about how tight convergence bounds are?",{"text":80,"@type":76},"The bounds are generally quite loose, and their relative magnitude reflects training loss more than testing loss.",{"name":82,"@type":73,"acceptedAnswer":83},"How can convergence-bound quantities help in federated learning without sharing data?",{"text":84,"@type":76},"Certain quantities in the bounds help identify clients most likely to contribute to learning while not requiring disclosure of dataset quality or size.","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},[93,97,101,105,109,114,119,122,127,130,134],{"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":21,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":21,"slug":137},19,"General","general"]