摘要
In Internet of Things (IoT), it is often impossible to share datasets owned by different participants (usually IoT devices) for machine learning model training due to privacy concerns. Federated learning (FL) is a promising technique to address this challenge. However, existing FL schemes face the problem of how to avoid low-quality/malicious update. To solve this problem, we propose a privacy-preserving and reliable federated learning scheme (PPRFLS) to select reliable participants and evaluate the quality of the participants' updates. Analysis shows that the proposed scheme achieves data privacy and model reliability.
学科与关键词
关键词
Federated learningModel reliabilityData privacy
- Privacy-Preserving and Reliable Decentralized Federated LearningGao, Yuanyuan; Zhang, Lei*; Wang, Lulu*; Choo, Kim-Kwang Raymond; Zhang, RuiIEEE Transactions on Services Computing, 2023, 16(4): 2879-2891.被引 1
- A Critical Evaluation of Privacy and Security Threats in Federated LearningAsad, Muhammad*; Moustafa, Ahmed; Yu, ChaoSENSORS, 2020, 20(24): 7182.被引 26
- Aggregation Service for Federated Learning: An Efficient, Secure, and More Resilient RealizationZheng, Yifeng*; Lai, Shangqi; Liu, Yi; Yuan, Xingliang; Yi, Xun; Wang, Cong*IEEE Transactions on Dependable and Secure Computing, 2023, 20(2): 988-1001.被引 22
