We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: Differentially Private SGD with Non-Smooth Loss

Abstract: In this paper, we are concerned with differentially private SGD algorithms in the setting of stochastic convex optimization (SCO). Most of existing work requires the loss to be Lipschitz continuous and strongly smooth, and the model parameter to be uniformly bounded. However, these assumptions are restrictive as many popular losses violate these conditions including the hinge loss for SVM, the absolute loss in robust regression, and even the least square loss in an unbounded domain. We significantly relax these restrictive assumptions and establish privacy and generalization (utility) guarantees for private SGD algorithms using output and gradient perturbations associated with non-smooth convex losses. Specifically, the loss function is relaxed to have $\alpha$-H\"{o}lder continuous gradient (referred to as $\alpha$-H\"{o}lder smoothness) which instantiates the Lipschitz continuity ($\alpha=0$) and strong smoothness ($\alpha=1$). We prove that noisy SGD with $\alpha$-H\"older smooth losses using gradient perturbation can guarantee $(\epsilon,\delta)$-differential privacy (DP) and attain optimal excess population risk $O\Big(\frac{\sqrt{d\log(1/\delta)}}{n\epsilon}+\frac{1}{\sqrt{n}}\Big)$, up to logarithmic terms, with gradient complexity (i.e. the total number of iterations) $T =O( n^{2-\alpha\over 1+\alpha}+ n).$ This shows an important trade-off between $\alpha$-H\"older smoothness of the loss and the computational complexity $T$ for private SGD with statistically optimal performance. In particular, our results indicate that $\alpha$-H\"older smoothness with $\alpha\ge {1/2}$ is sufficient to guarantee $(\epsilon,\delta)$-DP of noisy SGD algorithms while achieving optimal excess risk with linear gradient complexity $T = O(n).$
Comments: 26 pages, 1 table
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG)
MSC classes: 62P99
Cite as: arXiv:2101.08925 [stat.ML]
  (or arXiv:2101.08925v1 [stat.ML] for this version)

Submission history

From: Puyu Wang [view email]
[v1] Fri, 22 Jan 2021 03:19:06 GMT (50kb)
[v2] Sat, 26 Jun 2021 10:34:48 GMT (80kb,D)

Link back to: arXiv, form interface, contact.