Flamingo - Multi-Round Single-Server Secure Aggregation with Applications to Private Federated Learning

Explore a 16-minute IEEE conference talk on Flamingo, a multi-round single-server secure aggregation system designed for private federated learning applications. Delve into the research presented by Yiping Ma, Jess Woods, Sebastian Angel, Antigoni Polychroniadou, and Tal Rabin from the University of Pennsylvania, Microsoft Research, and JP Morgan AI Research. Learn about the innovative approach to secure aggregation in federated learning environments, focusing on privacy preservation and efficient data processing across multiple rounds using a single server architecture.