Yan Zhu

Federated Learning for Spatiotemporal Traffic Prediction: Privacy, Efficiency, Security and Emerging Extensions

This survey finds that distributed federated learning is a good way for privacy-preserving traffic prediction and needs more research on adaptive optimization, strong collaboration under mixed data, and combined privacy and security tools for real large-scale uses.