Federated Learning for Privacy-Preserving AI: Revolutionizing Data Sharing Across Industries

Abstract

Federated Learning (FL) has emerged as a groundbreaking approach to Artificial Intelligence (AI) that preserves user privacy while enabling collaborative model training across diverse datasets. This survey highlights the evolution, architecture, methodologies, and applications of FL in privacy-preserving data sharing across industries such as healthcare, finance, and edge computing. Challenges such as communication overhead, data heterogeneity, and security threats are discussed, alongside solutions leveraging encryption and differential privacy techniques. Real-world applications illustrate the transformative potential of FL in enabling secure, cross-enterprise AI.