PrivFusion: Privacy-Preserving Model Fusion via Decentralized Federated Graph Matching

doi:10.1109/TKDE.2024.3430819

	PrivFusion: Privacy-Preserving Model Fusion via Decentralized Federated Graph Matching
	Chen, Qian 1,2; Chen, Yiqiang 1,2,3; Jiang, Xinlong 1,2; Zhang, Teng 1,2; Dai, Weiwei 4; Huang, Wuliang 1,2; Yan, Bingjie 1,2; Yan, Zhen 1,2; Lu, Wang 5; Ye, Bo 6
	2024-12-01
发表期刊	IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
ISSN	1041-4347
卷号	36 期号:12 页码:9051-9064
摘要	Model fusion is becoming a crucial component in the context of model-as-a-service scenarios, enabling the delivery of high-quality model services to local users. However, this approach introduces privacy risks and imposes certain limitations on its applications. Ensuring secure model exchange and knowledge fusion among users becomes a significant challenge in this setting. To tackle this issue, we propose PrivFusion, a novel architecture that preserves privacy while facilitating model fusion under the constraints of local differential privacy. PrivFusion leverages a graph-based structure, enabling the fusion of models from multiple parties without additional training. By employing randomized mechanisms, PrivFusion ensures privacy guarantees throughout the fusion process. To enhance model privacy, our approach incorporates a hybrid local differentially private mechanism and decentralized federated graph matching, effectively protecting both activation values and weights. Additionally, we introduce a perturbation filter adapter to alleviate the impact of randomized noise, thereby recovering the utility of the fused model. Through extensive experiments conducted on diverse image datasets and real-world healthcare applications, we provide empirical evidence showcasing the effectiveness of PrivFusion in maintaining model performance while preserving privacy. Our contributions offer valuable insights and practical solutions for secure and collaborative data analysis within the domain of privacy-preserving model fusion.
关键词	Adaptation models Data models Privacy Load modeling Biological neural networks Collaboration Predictive models Graph matching model fusion model privacy
DOI	10.1109/TKDE.2024.3430819
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Plan of China[2023YFC3604802] ; Beijing Municipal Science & Technology Commission[Z221100002722009] ; National Natural Science Foundation of China[62202455] ; Youth Innovation Promotion Association CAS ; Science and technology innovation Program of Hunan Province[2022RC4006] ; Science and technology innovation Program of Hunan Province[2023WK2005] ; Science Research Foundation of the Joint Laboratory Project on Digital Ophthalmology and Vision Science[SZYK202201]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Computer Science, Information Systems ; Engineering, Electrical & Electronic
WOS记录号	WOS:001354743800120
出版者	IEEE COMPUTER SOC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/41193
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Chen, Yiqiang; Jiang, Xinlong
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing 100045, Peoples R China 2.Univ Chinese Acad Sci, Beijing 101408, Peoples R China 3.Peng Cheng Lab, Shenzhen 518066, Peoples R China 4.Changsha Aier Eye Hosp, Changsha 410015, Hunan, Peoples R China 5.Tsinghua Univ, Beijing 100190, Peoples R China 6.Nanchang Aier Eye Hosp, Nanchang 330002, Jiangxi, Peoples R China
推荐引用方式 GB/T 7714	Chen, Qian,Chen, Yiqiang,Jiang, Xinlong,et al. PrivFusion: Privacy-Preserving Model Fusion via Decentralized Federated Graph Matching[J]. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING,2024,36(12):9051-9064.
APA	Chen, Qian.,Chen, Yiqiang.,Jiang, Xinlong.,Zhang, Teng.,Dai, Weiwei.,...&Ye, Bo.(2024).PrivFusion: Privacy-Preserving Model Fusion via Decentralized Federated Graph Matching.IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING,36(12),9051-9064.
MLA	Chen, Qian,et al."PrivFusion: Privacy-Preserving Model Fusion via Decentralized Federated Graph Matching".IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 36.12(2024):9051-9064.