[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124919-en":3,"doc-seo-124919-105":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},124919,4398048950312,"Violet","https://ap-avatar.wpscdn.com/avatar/400002538284de19e3c?_k=1778320343897328908",6,"Technology","Towards Trustworthy Collaborative Machine Learning - A Blockchain-Driven Federated Approach","This thesis investigates how integrating Blockchain (BC) with Federated Learning (FL) can strengthen security and reliability in conventional FL systems. It reviews foundational principles of both FL and BC, then studies multiple Blockchain-Enabled Federated Learning (BCFL) architectures, analyzing their strengths and weaknesses. An implementation of one selected architecture is carried out to evaluate efficiency and effectiveness. Simulation results show significant BC computational overhead, yet model quality remains unaffected. Blockchain integrity is therefore presented as a promising mechanism to secure decentralized, distributed machine learning.","DIPARTIMENTO DI INGEGNERIA DELL’INFORMAZIONE  \nCORSO DI LAUREA IN INGEGNERIA DELL’INFORMAZIONE  \nTowards Trustworthy Collaborative Machine Learning: A Blockchain-Driven Federated  \nApproach  \nRelatore Laureanda  \nProf. Perin Giovanni Antonello Ilenia  \nANNO ACCADEMICO 2023-2024  \nData di laurea 19/03/2024  \nAbstract  \nThe purpose of this thesis is to explore how integrating Blockchain (BC) technology with Federated Learning (FL) can improve the security and reliability of standard FL systems. To achieve this, the principles of FL and BC are first reviewed. Next, various Blockchain Enabled Federated Learning (BCFL) architectures are theoretically examined, weighing their strengths and weaknesses. Finally, a straightforward implementation of one of the discussed architectures is conducted to assess its efficiency and efficacy. The simulation highlights that while the computational overhead associated with BC technology is significant, it does not affect the model’s quality. Despite this overhead, the integrity added to FL by BC makes it a promising solution for securing decentralized and distributed machine learning systems.  \nLo scopo di questa tesi è studiare come integrare la tecnologia Blockchain (BC) con il Federated Learning (FL) possa migliorare la sicurezza e l’affidabilità dei sistemi FL standard. Per raggiungere questo obiettivo, vengono prima esaminati i principi di FL e BC. Successivamente, varie architetture di Blockchain-Enabled Federated Learning (BCFL) vengonoesaminate teoricamente, valutandone i punti di forza e debolezze. Infine, viene condotta un’implementazione di una delle architetture discusse per valutarne l’efficienza e l’efficacia. La simulazione evidenzia che mentre il sovraccarico computazionale associato alla tecnologia BC è notevole, non ha effetto sulla qualità del modello. Nonostante questo sovraccarico, l’integrità aggiunta al FL dalla BC la rende una soluzione promettente per la sicurezza disistemi di apprendimento automatico decentralizzati e distribuiti.  \n4  \nContents  \n1 Introduction 7  \n1.1 Federated Learning ................................ 7  \n1.1.1 Principles ................................. 7  \n1.1.2 Advantages and challenges ........................ 9  \n1.2 Blockchain .................................... 9  \n1.2.1 Principles ................................. 9  \n1.2.2 Security Properties ............................ 11  \n2 Blockchain-Enabled Federated Learning 13  \n2.1 Integration of Blockchain and Federated Learning ................ 13  \n2.2 Advantages and challenges ............................ 15  \n3 Implementation 17  \n3.1 Description of the environment and tools used .................. 17  \n3.2 Overview of the Blockchain Implementation ................... 18  \n3.3 Overview of the Federated Learning Implementation .............. 22  \n3.4 Interaction between Blockchain and Federated Learning Structures ....... 23  \n4 Results 27  \n5 Conclusions 33  \n6  \nChapter 1  \nIntroduction  \n1.1 Federated Learning  \nIn today’s world, personal devices such as smartphones generate and store vast amounts of data. This data is vital in improving user experience by powering intelligent applications. However, the private nature of the data raises security concerns when stored remotely. Federated Learning (FL) was introduced to address these concerns.  \n1.1.1 Principles  \nFederated learning is a machine learning technique that sees a federation of users who collectively solve a training task coordinated by a server. Each client has its dataset, which is never uploaded but used to train a local model and compute an upgrade, which will be sent to the server. FL trains models using non-IID datasets that vary significantly in size depending on the device characteristics and usage.  \nIn this thesis, we consider the Federated Averaging (FedAvg) algorithm, which utilizes stochastic gradient descent (SGD) [2] . The traditional FL approach involves client selection, local model training using predefined algorit","cbCairKj8YOpUzVh","https://ap.wps.com/l/cbCairKj8YOpUzVh","pdf",752124,1,35,"English","en",105,"# Introduction\n## Federated Learning\n## Blockchain\n# Blockchain-Enabled Federated Learning\n## Integration of Blockchain and Federated Learning\n# Implementation\n## Description of the environment and tools used\n## Overview of the Blockchain Implementation\n## Overview of the Federated Learning Implementation\n## Interaction between Blockchain and Federated Learning Structures\n# Results\n# Conclusions","[{\"question\":\"How does Federated Learning protect data in the proposed approach?\",\"answer\":\"Federated Learning keeps user datasets local and does not upload raw data. Clients train local models and send only computed updates to the server for aggregation.\"},{\"question\":\"What problems in standard FL motivate the use of Blockchain?\",\"answer\":\"Standard FL faces risks such as single point of failure, lack of validity checks against poisoned updates, insufficient incentives, and user/device unreliability.\"},{\"question\":\"What is the impact of Blockchain on model quality according to the results?\",\"answer\":\"Simulations indicate that Blockchain introduces significant computational overhead, but it does not degrade the quality of the trained model.\"}]","Towards Trustworthy Collaborative Machine Learning - A Blockchain-Driven Federated Approach | PDF",1785895388,88,{"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},"towards-trustworthy-collaborative-machine-learning-a-blockchain-driven-federated-approach","",{"@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/technology/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/towards-trustworthy-collaborative-machine-learning-a-blockchain-driven-federated-approach/124919/",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},"How does Federated Learning protect data in the proposed approach?","Question",{"text":75,"@type":76},"Federated Learning keeps user datasets local and does not upload raw data. Clients train local models and send only computed updates to the server for aggregation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What problems in standard FL motivate the use of Blockchain?",{"text":80,"@type":76},"Standard FL faces risks such as single point of failure, lack of validity checks against poisoned updates, insufficient incentives, and user/device unreliability.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the impact of Blockchain on model quality according to the results?",{"text":84,"@type":76},"Simulations indicate that Blockchain introduces significant computational overhead, but it does not degrade the quality of the trained model.","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},[93,97,101,105,110,113,118,123,128,131,135],{"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":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":111,"slug":112},50,"technology",{"id":114,"doc_module":4,"doc_module_name":46,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},8,"Research & Report",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"]