Are Normalizing Flows the Key to Unlocking the Exponential Mechanism? A Path through the Accuracy-Privacy Ceiling Constraining Differentially Private ML

no code implementations • 15 Nov 2023 • Robert A. Bridges, Vandy J. Tombs, Christopher B. Stanley

The state of the art and de facto standard for differentially private machine learning (ML) is differentially private stochastic gradient descent (DPSGD).