A Review on Privacy Preserving Secure Machine Learning

Privacy preserving and secure machine learning (PPSML), has turned into increasingly significant in recent years due to the need to protect sensitive data while enabling data analysis and model training. The study explores various Privacy preserving techniques and evaluate their effectiveness in protecting sensitive data against security threats. The findings suggest that while many privacy preserving techniques can significantly improve data security, there is no common solution which fits all. Different techniques are better suited for different scenarios, depending on the type of attack, nature of the data, and the specific privacy requirements. Therefore, it is crucial to carefully evaluate and select the appropriate privacy preserving technique for each situation to ensure effective protection against security threats while maintaining data utility and performance.