[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125402-en":3,"doc-seo-125402-105":30,"detail-sidebar-cat-0-en-105":83},{"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},125402,1374391974564,"Clementine","https://ap-avatar.wpscdn.com/avatar/14000253aa45c000a9e?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779874745381141002",8,"Research & Report","Blind Evaluation Framework for Fully Homomorphic Encryption and Privacy-Preserving Machine Learning","Privacy-Preserving Machine Learning (PPML) commonly relies on Fully Homomorphic Encryption (FHE) to compute directly on encrypted data, yet encrypted training remains difficult because control structures require Interactive Rounds of Decryption and Evaluation (IRDE), also called client-assisted computation. In many outsourced settings, such client-server collaboration cannot be guaranteed, making existing approaches incompatible with practical deployment. This work proposes the Blind Evaluation Framework (BEF), enabling cryptographically secure encrypted control execution and removing IRDE during PPML training. An encrypted decision tree example achieves non-interactive training and improves IRDE efficiency by avoiding d-rounds for depth d, broadening FHE adoption where trusted clients cannot perform decryption rounds.","arXiv :2310 . 13140v4 [ cs .CR] 27 Aug 2024  \nReceived XX Month, XXXX; revised XX Month, XXXX; accepted XX Month, XXXX; Date of publication XX Month, XXXX; date of  \ncurrent version XX Month, XXXX.  \nDigital Object Identifier 10.1109/XXXX.2022.1234567  \nBlind Evaluation Framework for Fully Homomorphic Encryption and Privacy-Preserving Machine Learning  \nHunjae ”Timothy” Lee1 , Corey Clark1  \n1 Southern Methodist University, Dallas, TX 75205 USA  \nCorresponding author: Hunjae Lee (email: [hunjael@smu.edu](hunjael@smu.edu)).  \nThis work was funded by BALANCED Media|Technology (BMT), a company that may potentially benefit from the research results. Dr. Corey Clark has an equity interest in BMT and also serves as the company’s chief technology officer. The terms of this arrangement have been reviewed and approved by the Southern Methodist University in accordance with its conflict of interest policies.  \nABSTRACT In Privacy-Preserving Machine Learning (PPML), Fully Homomorphic Encryption (FHE) is often employed to enable computation directly on encrypted data. However, extensive programming with FHE have remained a challenge in large part owing to the difficulties in execution of control structures in encrypted state. While encrypted inference models are often immune from this issue due to their relative logical simplicity, training typically requires Interactive Rounds of Decryption and Evaluation (IRDE), also known as client-assisted computation, wherein certain operations are decrypted and evaluated in plaintext. However, such client-server communication with outsourced computing services cannot be reliably achieved in many scenarios such as in volunteer grids and distributed computing. These services have largely remained incompatible with PPML and FHE. To address this issue, we introduce the Blind Evaluation Framework (BEF), a cryptographically secure programming framework that enables encrypted execution of control structures and eliminates IRDE in training of PPML. To the best of our knowledge, we are the first to enable non-interactive training of PPML with FHE.  \nAs an example application of BEF, we implement an encrypted decision tree training model that doesn’t require IRDE, a drastic improvement from the previous state-of-the-art which required d-rounds of IRDE for tree-depth d. By advancing the state-of-the-art in IRDE efficiency by eliminating IRDE entirely, BEF enables adoption of FHE in use-cases where ample computing resources are available without the ability for trusted clients to perform decryption rounds.  \nINDEX TERMS Fully Homomorphic Encryption (FHE), Privacy-Preserving Machine Learning (PPML), Non-interactive training  \nI. INTRODUCTION  \nFULLY Homomorphic Encryption (FHE) is an encryp  \ntion scheme that allows computation on encrypted data without needing to decrypt them first, achieving encrypted computation without compromising cryptographic integrity [1] . Due to this property, FHE is frequently used in the research of secure outsourced computing and PrivacyPreserving Machine Learning (PPML) . However, there are limitations of FHE that create challenges for PPML. In [2]–[4], encrypted predictive models are deployed for privacypreserving neural networks but they are trained entirely in plaintext with FHE only being used for inference. In [5],[6], encrypted training of PPML is achieved but requires  \nthat the untrusted computing party (server) engage in Interactive Rounds of Decryption and Evaluation (IRDE) with the private-key owner (client) to handle some operations in decrypted, plaintext form to aid the server. This is also called the client-assisted computation model, referring to the fact that while PPML with FHE is about secure outsourced computation, the client is often needed in regular intervals to handle operations that are challenging to handle in encrypted form. In this work, IRDE and client-assisted computation are used interchangeably. A similar method is employed in [7], where a privacy-pr","cbCaingF5RU2D93e","https://ap.wps.com/l/cbCaingF5RU2D93e","pdf",632778,1,14,"English","en",105,"# Introduction\n## Fully Homomorphic Encryption (FHE)\n## Reality of PPML with FHE","[{\"question\":\"What improvement does BEF provide using an encrypted decision tree example?\",\"answer\":\"The decision tree training avoids d-rounds of IRDE for tree depth d, enabling non-interactive training and improving IRDE efficiency by eliminating IRDE entirely.\"}]","Blind Evaluation Framework for Fully Homomorphic Encryption and Privacy-Preserving Machine Learning | PDF",1785898692,35,{"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":78,"head_meta":80,"extra_data":82,"updated_unix":28},"blind-evaluation-framework-for-fully-homomorphic-encryption-and-privacy-preserving-machine-learning","",{"@graph":36,"@context":77},[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/blind-evaluation-framework-for-fully-homomorphic-encryption-and-privacy-preserving-machine-learning/125402/",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],{"name":72,"@type":73,"acceptedAnswer":74},"What improvement does BEF provide using an encrypted decision tree example?","Question",{"text":75,"@type":76},"The decision tree training avoids d-rounds of IRDE for tree depth d, enabling non-interactive training and improving IRDE efficiency by eliminating IRDE entirely.","Answer","https://schema.org",{"og:url":52,"og:type":79,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":81,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":84},[85,89,93,97,102,107,112,115,120,123,127],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":86,"show_sort_weight":87,"slug":88},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":90,"show_sort_weight":91,"slug":92},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Exam",70,"exam",{"id":98,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},5,"Comic",60,"comic",{"id":103,"doc_module":4,"doc_module_name":46,"category_name":104,"show_sort_weight":105,"slug":106},6,"Technology",50,"technology",{"id":108,"doc_module":4,"doc_module_name":46,"category_name":109,"show_sort_weight":110,"slug":111},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":113,"slug":114},30,"research-report",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":98,"slug":130},19,"General","general"]