AISTATS 2019poster19 citations

Renyi Differentially Private ERM for Smooth Objectives

Chen Chen, Jaewoo Lee, Dan Kifer

Abstract

In this paper, we present a Renyi Differentially Private stochastic gradient descent (SGD) algorithm for convex empirical risk minimization. The algorithm uses output perturbation and leverages randomness inside SGD, which creates a "randomized sensitivity", in order to reduce the amount of noise that is added. One of the benefits of output perturbation is that we can incorporate a periodic averaging step that serves to further reduce sensitivity while improving accuracy (reducing the well-known oscillating behavior of SGD near the optimum). Renyi Differential Privacy can be used to provide (epsilon, delta)-differential privacy guarantees and hence provide a comparison with prior work. An empirical evaluation demonstrates that the proposed method outperforms prior methods on differentially private ERM.

BibTeX
@InProceedings{pmlr-v89-chen19e,
  title = 	 {Renyi Differentially Private ERM for Smooth Objectives},
  author =       {Chen, Chen and Lee, Jaewoo and Kifer, Dan},
  booktitle = 	 {Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics},
  pages = 	 {2037--2046},
  year = 	 {2019},
  editor = 	 {Chaudhuri, Kamalika and Sugiyama, Masashi},
  volume = 	 {89},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {16--18 Apr},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v89/chen19e/chen19e.pdf},
  url = 	 {https://proceedings.mlr.press/v89/chen19e.html},
  abstract = 	 {In this paper, we present a Renyi Differentially Private stochastic gradient descent (SGD) algorithm for convex empirical risk minimization. The algorithm uses output perturbation and leverages randomness inside SGD, which creates a "randomized sensitivity", in order to reduce the amount of noise that is added. One of the benefits of output perturbation is that we can incorporate a periodic averaging step that  serves to further reduce sensitivity while improving accuracy (reducing the well-known oscillating behavior of SGD near the optimum).  Renyi Differential Privacy can be used to provide (epsilon, delta)-differential privacy guarantees and hence provide a comparison with prior work. An empirical evaluation demonstrates that the proposed method outperforms prior methods on differentially private ERM.}
}
Renyi Differentially Private ERM for Smooth Objectives · AISTATS 2019