A5052779809- Privacy-Preserving Technologies in Data
- Mobile Crowdsensing and Crowdsourcing
- Stochastic Gradient Optimization Techniques
- Privacy, Security, and Data Protection
Fuzhou University
2022-2024
Federated learning is a popular framework designed to perform the distributed machine while protecting client privacy. However, heterogeneous data distribution in real-world environments makes it difficult converge when performing model training. In this article, we propose federated gradient scheduling (FedGS), an improved historical sampling utilization method for optimizers that utilize gradients alleviate instability problem of information due non-IID. FedGS consists two main steps...
Centralized learning now faces data mapping and security constraints that make it difficult to carry out. Federated with a distributed architecture has changed this situation. By restricting the training process participants' local, federated addresses model needs of multiple sources while better protecting privacy. However, in real-world application scenarios, need achieve fairness addition privacy protection. In practice, could happen some participants specific motives may short join...
Federated learning is widely used and researched as an effective method for solving the privacy problems faced by centralized learning. To address communication limitations heterogeneity among clients, many existing methods based on mixup algorithm share data mixed with client's local dataset to improve model accuracy. However, due of federated learning, there may be some clients who join process insufficient data, which will violate privacy-preserving assumption mixup. Because this...