[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-118948-en":3,"doc-seo-118948-105":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},118948,3848291630094,"Emma Wilson","https://eur-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",6,"Technology","Practical Privacy-Preserving Machine Learning using Fully Homomorphic Encryption","Machine learning enables analysis of large datasets, yet privacy expectations and regulations restrict what sensitive information can be processed, even for societal benefit. Fully homomorphic encryption offers a way to compute directly on encrypted data, preserving privacy end-to-end, but its high computational cost has limited existing solutions mostly to encrypted inference rather than training. This work presents a practical approach for training using fully homomorphic encryption, delivering fast training speeds—under 45 seconds to train a binary classifier on thousands of samples—substantially outperforming prior state-of-the-art results.","Practical Privacy-Preserving Machine Learning using Fully  \nHomomorphic Encryption  \nMichael Brand  \nSchool of Computing Technologies, RMIT University Melbourne, Vic, Australia INCERT GIE Leudelange, Luxembourg [michael.brand@rmit.edu.au](michael.brand@rmit.edu.au)  \nGaëtan Pradel∗ Royal Holloway, University of London Information Security Group Egham, United Kingdom INCERT GIE Leudelange, Luxembourg [gpradel@incert.lu](gpradel@incert.lu)  \nABSTRACT  \nMachine learning is a widely-used tool for analysing large datasets, but increasing public demand for privacy preservation and the corresponding introduction of privacy regulations have severely limited what data can be analysed, even when this analysis is for societal benefit. Homomorphic encryption, which allows computation on encrypted data, is a natural solution to this dilemma, allowing data to be analysed without sacrificing privacy. Because homomorphic encryption is computationally expensive, however, current solutions are mainly restricted to use it for inference and not training.  \nIn this work, we present a practically viable approach to privacypreserving machine learning training using fully homomorphic encryption. Our method achieves fast training speeds, taking less than 45 seconds to train a binary classifier over thousands of samples on a single mid-range computer, significantly outperforming state-of-the-art results.  \nKEYWORDS  \nPrivacy, Fully Homomorphic Encryption, Machine Learning Training, Support Vector Machines  \n1 INTRODUCTION  \nIn recent years, Machine Learning (ML) techniques have become increasingly popular for analysing and extracting insights from large and complex datasets. With the rise of powerful computing systems and the availability of vast amounts of data from various domains, ML has found applications in numerous fields such as healthcare, finance, biometric recognition, surveillance and many others. As examples, ML is used in the medical field to analyse patient data to help with disease diagnosis and treatment, while in finance, it is used for fraud detection, risk assessment, and trading strategies [2, 48] .  \nHowever, the use of such techniques on sensitive data has raised significant concerns regarding data privacy [3, 36] . For example, medical data contains highly sensitive personal information and  \n∗ Corresponding author  \nThis work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license visit [https://creativecommons.org/licenses/by/4.0/ or](https://creativecommons.org/licenses/by/4.0/ or) send a  \nletter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA. Proceedings on Privacy Enhancing Technologies YYYY(X), 1–15  \n© YYYY Copyright held by the owner/author(s) .  \n[https://doi.org/XXXXXXX.XXXXXXX](https://doi.org/XXXXXXX.XXXXXXX)  \nfinancial data includes confidential transactional details that require protection. In response, privacy-related regulations such as the General Data Protection Regulation (GDPR) [22] in the European Union and the Health Insurance Portability and Accountability Act (HIPAA) [12] in the United States of America have been enacted, which require data holders to preserve the privacy of sensitive or personally-identifiable data.  \nNumerous techniques have been explored for enhancing privacy in data processing and analysis. These include cryptography, differential privacy, federated learning, and the use of secure enclaves [36, 56], with many tasks requiring a combination of such tools [39, 54] .  \nIn our work, we chose to tackle the privacy challenge using homomorphic encryption (HE) [40] . Introduced in 1978 by Rivest et al. [52], HE enables computations to be performed on encrypted data without requiring decryption, so that the privacy of the data can be preserved throughout the entire analysis process, i.e. both while the data is at rest and in use. Specifically, our goal was to use HE in order to enable privacy in Machine Learning, by introduci","cbCaijSxgCoGzkdG","https://ap.wps.com/l/cbCaijSxgCoGzkdG","pdf",856408,1,15,"English","en",105,"# Introduction\n## Privacy challenge in machine learning\n## Homomorphic encryption for encrypted computation\n## Prior work limitations: inference-only and computational cost\n## Focus of this paper: practical encrypted training with levelled FHE","[{\"question\":\"Why does privacy preservation limit machine learning on sensitive data?\",\"answer\":\"Sensitive data requires protection under privacy expectations and regulations, which can restrict what data can be analyzed even when the analysis benefits society.\"},{\"question\":\"How does fully homomorphic encryption address privacy in machine learning?\",\"answer\":\"Fully homomorphic encryption allows computation on encrypted data without decryption, preserving privacy during both data-at-rest and data-in-use phases.\"},{\"question\":\"What problem does this work target compared with existing solutions?\",\"answer\":\"It targets the practicality of the more computationally expensive training phase, whereas prior work largely focuses on encrypted inference with training performed in plaintext.\"}]","Practical Privacy-Preserving Machine Learning using Fully Homomorphic Encryption | PDF",1785721149,38,{"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},"practical-privacy-preserving-machine-learning-using-fully-homomorphic-encryption","",{"@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/practical-privacy-preserving-machine-learning-using-fully-homomorphic-encryption/118948/",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","2026-08-03",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 does privacy preservation limit machine learning on sensitive data?","Question",{"text":75,"@type":76},"Sensitive data requires protection under privacy expectations and regulations, which can restrict what data can be analyzed even when the analysis benefits society.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does fully homomorphic encryption address privacy in machine learning?",{"text":80,"@type":76},"Fully homomorphic encryption allows computation on encrypted data without decryption, preserving privacy during both data-at-rest and data-in-use phases.",{"name":82,"@type":73,"acceptedAnswer":83},"What problem does this work target compared with existing solutions?",{"text":84,"@type":76},"It targets the practicality of the more computationally expensive training phase, whereas prior work largely focuses on encrypted inference with training performed in plaintext.","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"]