Download PDFOpen PDF in browserProtecting Machine Learning Insights: Ensuring Data Privacy with Advanced Privacy-Preserving TechniquesEasyChair Preprint 124388 pages•Date: March 10, 2024AbstractWith the increasing reliance on machine learning (ML) for extracting valuable insights from data, the need to safeguard sensitive information has become paramount. This paper explores advanced privacy-preserving techniques aimed at securing ML insights and ensuring data privacy. We delve into cryptographic methods, federated learning, and differential privacy, offering a comprehensive overview of their applications in the ML landscape. Cryptographic techniques such as homomorphic encryption enable computations on encrypted data, ensuring that sensitive information remains confidential throughout the ML process. Federated learning facilitates model training across decentralized devices without centralizing raw data, preserving user privacy. Differential privacy introduces noise to individual data points, striking a balance between accurate model training and safeguarding individual contributions. Keyphrases: Federated Learning, Privacy-preserving techniques, cryptographic methods, data privacy, differential privacy, machine learning
|