Skip to main content

🏢 Texas State University

Differentially Private Stochastic Gradient Descent with Fixed-Size Minibatches: Tighter RDP Guarantees with or without Replacement
·2045 words·10 mins· loading · loading
AI Theory Privacy 🏢 Texas State University
Tighter differential privacy (RDP) guarantees for DP-SGD with fixed-size minibatches are achieved, improving private deep learning model training.