In previous chapters we’ve looked into differential privacy, local differential privacy, privacy-preserving synthetic data generation, privacy-preserving data mining, and their application when designing privacy-preserving machine learning solutions. As you’ll recall, in differential privacy a trusted data curator collects data from individuals and produces differentially private results by adding precisely computed noise to the aggregation of individuals’ data. In local differential privacy, individuals privatize their own data by perturbation before sending it to the data aggregator, which eliminates the need to have a trusted data curator collect the data from individuals. In data mining, we looked into various privacy-preserving techniques and operations that can be used when collecting information and publishing the data. We also discussed strategies for regulating data mining output. Privacy- preserving synthetic data generation provides a promising solution for private data sharing, where synthetic yet representative data can be generated and then shared among multiple parties safely and securely.
