We consider training models with differential privacy (DP) using mini-batch gradients. The existing state-of-the-art, Differentially Private Stochastic Gradient Descent (DP-SGD), requires privacy amplification by sampling or shuffling to obtain the best privacy/accuracy/computation trade-offs. Unfortunately, the precise requirements on exact sampling and shuffling can be hard to obtain in important practical scenarios, particularly federated learning (FL). We design and analyze a DP variant of Follow-The-Regularized-Leader (DP-FTRL) that compares favorably (both theoretically and empirically) to amplified DP-SGD, while allowing for much more flexible data access patterns. DP-FTRL does not use any form of privacy amplification. The code is available at https://github.com/google-research/federated/tree/master/dp_ftrl and https://github.com/google-research/DP-FTRL .
@article{arxiv.2103.00039,
title = {Practical and Private (Deep) Learning without Sampling or Shuffling},
author = {Peter Kairouz and Brendan McMahan and Shuang Song and Om Thakkar and Abhradeep Thakurta and Zheng Xu},
journal= {arXiv preprint arXiv:2103.00039},
year = {2021}
}