Federated Learning in Cloud–Edge Environments for Privacy-Preserving Cognitive Computing

Abstract

Federated learning (FL) enables collaborative machine learning without transferring raw data to a central server, thereby ensuring privacy and security. When integrated with cloud–edge environments, FL enhances cognitive computing by enabling real-time, decentralized intelligence. This paper explores the architecture, opportunities, applications, and challenges of federated learning for privacy-preserving cognitive systems. It highlights how cloud–edge collaboration improves data security, latency, scalability, and model performance while addressing integration barriers, communication overhead, and ethical concerns.