Year: 2025
Type: article
Abstract: Federated Learning (FL) has emerged as a cutting-edge technique for privacy-preserving machine learning in distributed systems. Unlike traditional machine learning, which relies on centralized data storage, FL enables model training directly on decentralized data sources, ensuring that sensitive information never leaves its local environment. This paper explores the integration of Federated Learning with AI-powered privacy frameworks, focusing on secure multi-party computation, differential privacy, and cryptographic techniques to further safeguard user data. Through a comprehensive review of existing FL models and privacy-enhancing methods, the paper discusses how federated learning can be leveraged to address the challenges of data security, user privacy, and computational efficiency in distributed systems, particularly in fields like healthcare, finance, and IoT. The proposed framework demonstrates how Federated Learning, combined with AI-driven privacy techniques, can foster more trustworthy and secure collaborative machine learning processes while minimizing data leakage risks.
Authors: Prudhivi Anuradha, C. Arunbala, U. Harita, K. Valarmathi, S. Thenappan, V. Saravanan
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4411517184