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.
翻译:通过Stochatic Gradient Langevin Dynamics(SGLD)进行贝叶斯学习是建议进行差别化私人学习的。虽然以前的研究为SGLD在算法初始阶段或接近趋同时规定了不同的隐私界限,但是在两者之间可以作出什么不同的隐私保障的问题仍然没有答案。这个临时区域非常重要,特别是对Bayesian神经网络来说,因为很难保证与后方的趋同。 本文表明,使用SGLD可能会给这个临时区域造成不受限制的隐私损失,即使从后方取样的隐私与预期的一样不同。