Privacy-Preserving Federated Machine Learning Techniques
Gobinath Subramaniam, Santhiya Palanisamy · Advances in information security, privacy, and ethics book series · 2023
Machine learning is increasingly used for data analysis, but centralized datasets raise concerns about data privacy and security. Federated learning, a distributed method, enables multiple entities to cooperatively train a machine learning model. Clients use their local datasets to train local models, while a central aggregator aggregates updates and computes a global model. Privacy-preserving federated learning (PPFL) addresses privacy issues in sensitive and decentralized data situations. PPFL integrates federated learning with privacy-preserving approaches to achieve both privacy and model correctness.