About this document
Personalization Improves Privacy-Accuracy Tradeoffs in Federated Learning by ratneshwar.net is a document available to read on EtoBox.
This paper explores the tradeoff between privacy and accuracy in personalized federated learning by introducing a personalization parameter that balances local and global model learning. It demonstrates that adjusting this parameter can enhance generalization while maintaining user-level differential privacy. The authors provide theoretical guarantees and empirical results to support their findings, showing improved accuracy and privacy in federated learning scenarios.
- Author
- ratneshwar.net
- Language
- EN