Tangent differential privacy
This work addresses privacy protection in machine learning by offering a distribution-specific alternative to uniform differential privacy, though it appears incremental as it builds on existing differential privacy concepts.
The paper introduces tangent differential privacy, a new framework tailored to specific data distributions and allowing general distribution distances, and shows that entropic regularization guarantees it under general conditions for risk minimization.
Differential privacy is a framework for protecting the identity of individual data points in the decision-making process. In this note, we propose a new form of differential privacy called tangent differential privacy. Compared with the usual differential privacy that is defined uniformly across data distributions, tangent differential privacy is tailored towards a specific data distribution of interest. It also allows for general distribution distances such as total variation distance and Wasserstein distance. In the case of risk minimization, we show that entropic regularization guarantees tangent differential privacy under rather general conditions on the risk function.