Can Stochastic Gradient Langevin Dynamics Provide Differential Privacy for Deep Learning?Download PDF

24 Aug 2022 (modified: 05 May 2023)SaTML 2023Readers: Everyone
Keywords: differential privacy, bayesian inference, SGLD
TL;DR: The paper investigates, theoretically and empirically, whether SGLD can provide differential privacy for deep learning.
Abstract: Bayesian learning via Stochastic Gradient Langevin Dynamics (SGLD) has been suggested for differentially private learning. While previous research provides differential privacy bounds for SGLD at the initial steps of the algorithm or when close to convergence, the question of what differential privacy guarantees can be made in between remains unanswered. This interim region is of great importance, especially for Bayesian neural networks, as it is hard to guarantee convergence to the posterior. This paper shows that using SGLD might result in unbounded privacy loss for this interim region, even when sampling from the posterior is as differentially private as desired.
0 Replies