[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119993-en":3,"doc-seo-119993-105":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},119993,137441390410,"Hazel","https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984",8,"Research & Report","The High-Level Practical Overview of Open-Source Privacy-Preserving Machine Learning Solutions - Practical comparison","The paper presents a high-level, practical overview of privacy-preserving machine learning approaches that protect customer confidentiality. It evaluates offline-learning security methods, including homomorphic encryption, secure multi-party computation, and trusted execution environments such as Intel SGX, then studies their integration with different ML architectures via a proof of concept. The work compares open-source Python PPML solutions—SyMPC, TF-Encrypted, TenSEAL, and Gramine—using MNIST image classification and reports similar secure accuracy across methods, while highlighting large overheads versus non-secure evaluation.","INTL JOURNAL OF ELECTRONICS AND TELECOMMUNICATIONS, 2022, VOL. 68, NO. 4, PP. 741–747  \nManuscript received September 2, 2022; revised November, 2022 . DOI: 10.24425/ijet.2022.143880  \nThe High-Level Practical Overview of Open-Source Privacy-Preserving Machine Learning Solutions  \nKonrad Ku´zniewski, Krystian Matusiewicz, Piotr Sapiecha  \nAbstract—This paper aims to provide a high-level overview of practical approaches to machine-learning respecting the privacy and confidentiality of customer information, which is called Privacy-Preserving Machine Learning. First, the security approaches in offline-learning privacy methods are assessed. Those focused on modern cryptographic methods, such as Homomorphic Encryption and Secure Multi-Party Computation, as well as on dedicated combined hardware and software platforms like Trusted Execution Environment - Intel® Software Guard Extensions (Intel® SGX). Combining the security approaches with different machine learning architectures leads to our Proof of Concept in which the accuracy and speed of the security solutions will be examined. The next step was exploring and comparing the Open-Source Python-based solutions for PPML. Four solutions were selected from almost 40 separate, state-of-the-art systems: SyMPC, TF-Encrypted, TenSEAL, and Gramine. Three different Neural Network architectures were designed to show different libraries’ capabilities. The POC solves the image classification problem based on the MNIST dataset. As the computational results show, the accuracy of all considered secure approaches is similar. The maximum difference between non-secure and secure flow does not exceed 1.2% . In terms of secure computations, the most effective Privacy-Preserving Machine Learning library is based on Trusted Execution Environment, followed by Secure Multi-Party Computation and Homomorphic Encryption. However, most of those are at least 1000 times slower than the nonsecure evaluation. Unfortunately, it is not acceptable for a realworld scenario. Future work could combine different security approaches, explore other new and existing state-of-the-art libraries or implement support for hardware-accelerated secure computation.  \nKeywords—Privacy-Preserving Machine Learning, Homomorphic Encryption, Secure Multi Party Computation, Trusted Execution Environment  \nI. INTRODUCTION  \nRECENT advances in Machine Learning (ML) or Deep  \nLearning (DL) techniques have demonstrated outstanding performance on various tasks, including organ recognition from medical images, classification of interstitial lung diseases, detection of lung nodules, medical image reconstruction, and segmentation of brain tumors. The advantage of DL models over humans has resulted in the development of computeraided diagnosis systems-for example, the United States Food and Drug Administration recently announced the approval of an intelligent diagnosis system for medical images that do not require human intervention [1], [2] .  \nKonrad Ku´zniewski, Krystian Matusiewicz, Piotr Sapiechaare with Intel, the IPAS division (e-mail: {konrad.kuzniewski, [krystian.matusiewicz](krystian.matusiewicz}@intel.com)[}](krystian.matusiewicz}@intel.com)[@intel.com](krystian.matusiewicz}@intel.com), [piotr.sapiecha@intel.com](piotr.sapiecha@intel.com)).  \nNowadays, deep model training and evaluation are frequently outsourced to clouds, referred to in the literature as Machine Learning as a Service (MLaaS) . Cloud providers like Google, Microsoft Azure, or Amazon Web Services offer these services. Despite the impressive performance of DL algorithms, numerous recent studies have raised concerns about the security and robustness of machine learning models [3]–[5] . Moreover, the security of such algorithms’ execution environments is being questioned [6]–[8] . The realization that DL models are neither safe nor resilient considerably complicates their practical implementation in security-critical applications such as predictive healthcare, which is ba","cbCaioBMBMM9KOVL","https://ap.wps.com/l/cbCaioBMBMM9KOVL","pdf",854001,1,7,"English","en",105,"# I. Introduction\n## II. Security Overview for Machine Learning\n## Privacy-Preserving Machine Learning Security Approaches\n## Open-Source PPML Solutions Comparison","[{\"question\":\"What security approaches are assessed for privacy-preserving machine learning?\",\"answer\":\"The paper assesses offline-learning privacy methods based on modern cryptography, including homomorphic encryption and secure multi-party computation, as well as trusted execution environments such as Intel SGX.\"},{\"question\":\"Which open-source Python PPML solutions are compared in the proof of concept?\",\"answer\":\"SyMPC, TF-Encrypted, TenSEAL, and Gramine are selected from nearly 40 state-of-the-art systems for comparison.\"},{\"question\":\"How do secure approaches compare to non-secure evaluation in performance and accuracy?\",\"answer\":\"The accuracy differences among secure approaches are similar, and the maximum gap between non-secure and secure flow does not exceed 1.2%, but secure computations are at least 1000 times slower than non-secure evaluation.\"}]","The High-Level Practical Overview of Open-Source Privacy-Preserving Machine Learning Solutions - Practical comparison | PDF",1785727562,18,{"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},"the-high-level-practical-overview-of-open-source-privacy-preserving-machine-learning-solutions-practical-comparison","",{"@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/the-high-level-practical-overview-of-open-source-privacy-preserving-machine-learning-solutions-practical-comparison/119993/",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},"What security approaches are assessed for privacy-preserving machine learning?","Question",{"text":75,"@type":76},"The paper assesses offline-learning privacy methods based on modern cryptography, including homomorphic encryption and secure multi-party computation, as well as trusted execution environments such as Intel SGX.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which open-source Python PPML solutions are compared in the proof of concept?",{"text":80,"@type":76},"SyMPC, TF-Encrypted, TenSEAL, and Gramine are selected from nearly 40 state-of-the-art systems for comparison.",{"name":82,"@type":73,"acceptedAnswer":83},"How do secure approaches compare to non-secure evaluation in performance and accuracy?",{"text":84,"@type":76},"The accuracy differences among secure approaches are similar, and the maximum gap between non-secure and secure flow does not exceed 1.2%, but secure computations are at least 1000 times slower than non-secure evaluation.","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},[93,97,101,105,110,115,119,122,127,130,134],{"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":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"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":106,"slug":137},19,"General","general"]