[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124001-en":3,"doc-seo-124001-105":30,"detail-sidebar-cat-0-en-105":95},{"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},124001,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Machine Learning Techniques for the Preservation of Data Privacy - Dissertation","Machine learning often requires training and testing data that may contain sensitive information, making sharing with model owners infeasible under privacy requirements. This dissertation addresses privacy-preserving ML when encrypted computation and federated learning are necessary but still introduce risks during training, model distribution, and inference. It provides background on learning with encryption and federated learning, then proposes novel private inference methods that protect both the model and the inference data. A federated learning framework is introduced to preserve training data privacy and enable a single participant to obtain the jointly trained model. The work also presents secure distributed dimensionality reduction and combines everything into an end-to-end private framework that protects federated learning, inference privacy, and trained parameter confidentiality.","Machine Learning Techniques for the Preservation of Data Privacy  \nBy  \nWilliam R. Briguglio  \nA Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of  \nDoctor of Philosophy (Ph.D.)  \nin the Department of Electrical and Computer Engineering  \n©William R. Briguglio, 2024  \nUniversity of Victoria  \nAll rights reserved. This dissertation may not be reproduced in whole or in part, by photocopy or other means, without the permission of the author.  \nMachine Learning Techniques for the Preservation of Data Privacy  \nby  \nWilliam R. Briguglio  \nB.C.S. , University of Windsor – 2019  \nM.Sc. , University of Windsor – 2020  \nSupervisory Committee:  \n\n| Dr. Issa Traor´e, Co-Supervisor\u003Cbr>(Department of Electrical and Computer Engineering) |\n| --- |\n| Dr. Waleed Yousef, Co-supervisor\u003Cbr>(Department of Electrical and Computer Engineering) |\n| Dr. Sherif Saad, Departmental Member\u003Cbr>(Department of Electrical and Computer Engineering) |\n\nDr. Hausi Muller, Outside Member (Department of Computer Science)  \nAbstract  \nMachine learning has been successfully applied in various domains in recent years. Still, its use is limited since training or testing data may contain sensitive information which cannot be shared with model owners due to privacy concerns. For example, healthcare providers may be bound by patient privacy laws. Without large publicly available datasets, useful models become impossible to train. Therefore, ML methods that preserve the privacy of private training data are required. One solution is to use homomorphic encryption to carryout mathematical operations on encrypted data without compromising the privacy of said data. However, sometimes a large dataset that is difficult for a single institution to obtain is needed for complex learning tasks. In such a case, federated learning can be used to learn from private data distributed across multiple owners without compromising the privacy of each owner’s data.  \nHowever, federated learning carries its own risks. For example, exchanging even the minimum information needed for training can compromise privacy, and rogue participants in a federated learning network may attempt to sabotage model performance. Further, data that is not independently and identically distributed hampers the convergence of federated learning techniques. Additionally, once training is complete, regardless of the means, extra steps must be taken to ensure model privacy during the inference phase. Such steps are needed to ensure the model owner(s) can retain sole proprietorship of the global model. Further, if a model’s parameters are leaked, then an adversary may be able to reverse engineer them to compromise the privacy of the training data. Keeping a model private eliminates this risk.  \nIn this dissertation, we provide an in-depth background to the problems of machine learning with encryption and federated learning. We propose novel techniques for private inference that maintain the privacy of both the model and the data the model performs inferences on. We propose a federated learning framework which, in addition to maintaining the privacy of the data used during training, is, to the best of our knowledge, the only approach that enables just a single participant to obtain the jointly trained model. We also present a secure method for distributed dimensionality reduction, which can be used as apreprocessing step to enhance the performance of the proposed federated learning framework. Finally, we combine these approaches and propose an end-to-end federated learning and private inference framework which maintains data privacy during the federated learning and private inference phase, as well as ensures the privacy of the trained model’s parameters during each phase.  \nContents  \nSupervisory Committee ii  \nAbstract iii  \nContents iv  \nList of Tables vii  \nList of Figures viii  \nList of Algorithms x  \nList of Acronyms xi  \nList of Appendices xiii  \nAcknowledgment xiv  \nDedication xv  \n1","cbCaidV3iUfYpxL7","https://ap.wps.com/l/cbCaidV3iUfYpxL7","pdf",1688254,1,144,"English","en",105,"# 1 Introduction\n## 1.1 Problem Statement\n## 1.2 Research Contributions\n## 1.3 List of Publications\n## 1.4 Report Outline\n# 2 Background and Related Work\n## 2.0.1 Data Flow and Threat Model\n## 2.1 Private Machine Learning\n## 2.1.1 Homomorphic Encryption\n## 2.1.2 Machine Learning with Encryption\n## 2.2 Federated Learning\n## 2.2.1 Federated Dimensionality Reduction\n# 3 Machine Learning via Encryption\n## 3.1 A Machine Learning with Encryption (MLE) Framework\n## 3.2 Experiments\n## 3.2.1 MSK-IMPACT Dataset\n## 3.2.2 Building the Model\n## 3.2.3 Encrypting the Testing Dataset\n## 3.2.4 MLE Framework: Opensource and Deployment\n## 3.2.5 Discussion\n## 3.3 Summary\n# 4 FLAMED\n## 4.1 Method\n## 4.1.1 FLAMED: The General Framework\n## 4.1.2 FLAMED: Practical Benefits\n## 4.1.3 FLAMED: Specific Implementation and Parameters\n## 4.1.4 Comparison Approaches","[{\"question\":\"Why are privacy-preserving machine learning methods required?\",\"answer\":\"Training or testing data may contain sensitive information that cannot be shared due to privacy concerns and legal obligations. Large public datasets may be unavailable, preventing useful model training without privacy-preserving methods.\"},{\"question\":\"How do homomorphic encryption and federated learning help protect privacy?\",\"answer\":\"Homomorphic encryption enables computations on encrypted data without revealing underlying values. Federated learning learns from private data distributed across multiple owners while avoiding direct sharing of each owner’s raw data.\"},{\"question\":\"What additional risks exist in federated learning and after training?\",\"answer\":\"Federated learning can leak information through exchanged training signals and can be disrupted by rogue participants. Even after training, inference-stage privacy and parameter privacy must be protected to prevent adversaries from compromising training data.\"},{\"question\":\"What does the dissertation propose to maintain privacy during training and inference?\",\"answer\":\"It proposes private inference techniques that keep both the model and inference data confidential. It also introduces a federated learning framework enabling a single participant to obtain the jointly trained model, plus secure distributed dimensionality reduction, and combines them into an end-to-end system protecting privacy across phases.\"}]","Machine Learning Techniques for the Preservation of Data Privacy - Dissertation | PDF",1785819749,363,{"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":90,"head_meta":92,"extra_data":94,"updated_unix":28},"machine-learning-techniques-for-the-preservation-of-data-privacy-dissertation","",{"@graph":36,"@context":89},[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/machine-learning-techniques-for-the-preservation-of-data-privacy-dissertation/124001/",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","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81,85],{"name":72,"@type":73,"acceptedAnswer":74},"Why are privacy-preserving machine learning methods required?","Question",{"text":75,"@type":76},"Training or testing data may contain sensitive information that cannot be shared due to privacy concerns and legal obligations. Large public datasets may be unavailable, preventing useful model training without privacy-preserving methods.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How do homomorphic encryption and federated learning help protect privacy?",{"text":80,"@type":76},"Homomorphic encryption enables computations on encrypted data without revealing underlying values. Federated learning learns from private data distributed across multiple owners while avoiding direct sharing of each owner’s raw data.",{"name":82,"@type":73,"acceptedAnswer":83},"What additional risks exist in federated learning and after training?",{"text":84,"@type":76},"Federated learning can leak information through exchanged training signals and can be disrupted by rogue participants. Even after training, inference-stage privacy and parameter privacy must be protected to prevent adversaries from compromising training data.",{"name":86,"@type":73,"acceptedAnswer":87},"What does the dissertation propose to maintain privacy during training and inference?",{"text":88,"@type":76},"It proposes private inference techniques that keep both the model and inference data confidential. It also introduces a federated learning framework enabling a single participant to obtain the jointly trained model, plus secure distributed dimensionality reduction, and combines them into an end-to-end system protecting privacy across phases.","https://schema.org",{"og:url":52,"og:type":91,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":93,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":96},[97,101,105,109,114,119,124,127,132,135,139],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},"Exam",70,"exam",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},5,"Comic",60,"comic",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},6,"Technology",50,"technology",{"id":120,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":122,"slug":123},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":125,"slug":126},30,"research-report",{"id":128,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":130,"slug":131},9,"Religion & Spirituality",20,"religion-spirituality",{"id":130,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":130,"slug":134},"World Cup","world-cup",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":136,"slug":138},10,"Lifestyle","lifestyle",{"id":140,"doc_module":4,"doc_module_name":46,"category_name":141,"show_sort_weight":110,"slug":142},19,"General","general"]