AI RESEARCH
Trade-off Functions for DP-SGD with Subsampling based on Random Shuffling: Tight Upper and Lower Bounds
arXiv CS.LG
•
ArXi:2605.06259v1 Announce Type: new We derive a tight analysis of the trade-off function for Differentially Private Stochastic Gradient Descent (DP-SGD) with subsampling based on random shuffling within the $f$-DP framework. Our analysis covers the regime $\sigma \geq \sqrt{3/\ln M}$, where $\sigma$ is the noise multiplier and $M$ is the number of rounds within a single epoch.