[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126801-en":3,"doc-seo-126801-105":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":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},126801,1099523882182,"Eliana","https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8",8,"Research & Report","Privacy Attacks and Protection in Generative Models - Dissertation","Recent years have seen rapid progress in generative models for data synthesis, where the trained model and its training set become critical assets for model owners. This dissertation addresses privacy risks in practice: exposure of model privacy can undermine intellectual property and market position, while disclosure of training data, especially personal information, can violate data-privacy laws and trigger sanctions. New attacks and defenses are developed for model privacy and data privacy, along with a generative-model-based risk analysis approach for discriminative models.","PhD-FSTM-2023-130  \nThe Faculty of Science, Technology and Medicine  \nDISSERTATION  \nDefence held on 12/12/2023 in Esch-sur-Alzette  \nto obtain the degree of  \nDOCTEUR DE L’UNIVERSITÉ DU LUXEMBOURG  \nEN INFORMATIQUE  \nby  \nHailong HU  \nBorn on 15 April 1994 in Sichuan, (China)  \nPRIVACY ATTACKS  \nAND PROTECTION IN GENERATIVE MODELS  \nDissertation defence committee Dr Jun PANG, dissertation supervisor Professor, Université du Luxembourg  \nDr Mathias HUMBERT  \nProfessor, University of Lausanne  \nDr Gabriele LENZINI, Chairman Professor, Université du Luxembourg  \nDr Yang ZHANG, Vice Chairman  \nProfessor, CISPA Helmholtz Center for Information Security  \niii  \nThis dissertation was funded in whole by the Luxembourg National Research Fund (FNR)  \nGrant reference 13550291  \nv  \nAbstract  \nPrivacy Attacks and Protection in Generative Models  \nby Hailong Hu  \nRecent years have witnessed the tremendous success of generative models in data synthesis. Typically, a well-trained model itself and its training set constitute key assets for model owners, which allows technology companies to gain a leading position in the global market. However, privacy is a key consideration in deploying state-of-the-art generative models in practice. On the one hand, the exposure of model privacy can lead to the compromise of the intellectual property rights of legitimate model owners, which consequently affects the market share of companies. On the other hand, the disclosure of training data, especially when it includes personal information, constitutes a direct infringement of data privacy, which severely leads to legal sanctions for companies. Indeed, the advent of emerging generative models critically necessitates novel privacy analysis and protection techniques to ensure the confidentiality of cutting-edge models and their training data. To solve these challenges, this dissertation investigates several new privacy attacks and protection methods for generative models from the perspective of model privacy and data privacy. In addition, this dissertation also explores a new mode that leverages existing pre-trained generative models to study the security vulnerabilities of discriminative models, which provides a fresh angle to apply generative models to the risk analysis of discriminative models.  \nThis dissertation is organized into three parts. In the first part, i.e. model privacy in generative models, I develop new model extraction attacks to steal generative adversarial networks (GANs) . The evaluations show that preventing model extraction attacks against GANs is difficult but protecting GANs through verifying the ownership can be a deterrence against malicious adversaries. Thus, I further propose an ownership protection method to safeguard GANs, which can effectively recognize these stolen models constructed from physical stealing and model extraction. In the second part, i.e. data privacy in generative models, I develop two types of membership inference attacks against diffusion models, and the proposed lossbased method reveals the relationship between membership inference risks and the generative mechanism of diffusion models. I also investigate property inference risks in diffusion models and propose the first property aware sampling method to mitigate this attack, which bears the benefits of being plug-in and model-agnostic. In the third part, i.e. applications of generative models, I propose a new typeof out-of-distribution (OOD) attack by leveraging off-the-shelf pre-trained GANs, which demonstrates that GANs can be utilized to directly construct samples to fool classification models and evade OOD detection. Taken together, this dissertation primarily provides new privacy attacks and protection methods for generative models and can contribute to a deeper and more comprehensive understanding of the privacy of generative artificial intelligence.  \nvii  \nAcknowledgements  \nThe Ph.D. journey at the University of Luxembourg is a precious chapter in ","cbCaig066FvR37Oy","https://ap.wps.com/l/cbCaig066FvR37Oy","pdf",5618081,1,181,"English","en",105,"# Introduction\n## Generative Models\n## Privacy Attacks and Protection\n## Dissertation Structure","[{\"question\":\"What privacy threats does the dissertation focus on in generative models?\",\"answer\":\"It focuses on threats to model privacy (e.g., model extraction) and data privacy (e.g., membership inference and property inference) in generative models.\"},{\"question\":\"What defenses are proposed for model privacy, especially against GAN extraction?\",\"answer\":\"The dissertation proposes ownership protection to deter adversaries by verifying and recognizing stolen models constructed through physical stealing and model extraction.\"},{\"question\":\"How are membership inference risks studied for diffusion models?\",\"answer\":\"It develops two types of membership inference attacks against diffusion models and uses a loss-based method to reveal how membership inference risks relate to the diffusion generative mechanism.\"},{\"question\":\"What is the out-of-distribution attack contribution using pre-trained GANs?\",\"answer\":\"It proposes an out-of-distribution attack that leverages off-the-shelf pre-trained GANs to generate samples that can fool classification models and evade OOD detection.\"}]","Privacy Attacks and Protection in Generative Models - Dissertation | PDF",1785934869,456,{"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},"privacy-attacks-and-protection-in-generative-models-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/privacy-attacks-and-protection-in-generative-models-dissertation/126801/",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",[71,77,81,85],{"name":72,"@type":73,"acceptedAnswer":74},"What privacy threats does the dissertation focus on in generative models?","Question",{"text":75,"@type":76},"It focuses on threats to model privacy (e.g., model extraction) and data privacy (e.g., membership inference and property inference) in generative models.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What defenses are proposed for model privacy, especially against GAN extraction?",{"text":80,"@type":76},"The dissertation proposes ownership protection to deter adversaries by verifying and recognizing stolen models constructed through physical stealing and model extraction.",{"name":82,"@type":73,"acceptedAnswer":83},"How are membership inference risks studied for diffusion models?",{"text":84,"@type":76},"It develops two types of membership inference attacks against diffusion models and uses a loss-based method to reveal how membership inference risks relate to the diffusion generative mechanism.",{"name":86,"@type":73,"acceptedAnswer":87},"What is the out-of-distribution attack contribution using pre-trained GANs?",{"text":88,"@type":76},"It proposes an out-of-distribution attack that leverages off-the-shelf pre-trained GANs to generate samples that can fool classification models and evade OOD detection.","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"]