Private Hypothesis Selection

By Mark Bun, Gautam Kamath, Thomas Steinke, Zhiwei Steven Wu. 2019

101 citations recorded by the source corpus.

We provide a differentially private algorithm for hypothesis selection. Given samples from an unknown probability distribution P and a set of m probability distributions <inline-formula> <tex-math notation="LaTeX">$\mathcal {H}$ </tex-math></inline-formula>, the goal is to output, in a <inline-formula> <tex-math notation="LaTeX">$\varepsilon $ </tex-math></inline-formula>-differentially private manner, a distribution from <inline-formula> <tex-math notation="LaTeX">$\mathcal {H}$ </tex-math></inline-formula> whose total variation distance to P is comparable to that of the best such distribution (which we denote by <inline-formula> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula>). The sample complexity of our basic algorithm is <inline-formula> <tex-math notation="LaTeX">$\text {O}\left ({\frac {\log \text {m}}{\alpha ^{2}} + \frac {\log \text {m}}{\alpha \varepsilon }}\right)$ </tex-math></inline-formula>, representing a minimal cost for privacy when compared to the non-private algorithm. We also can handle infinite hypothesis classes <inline-formula> <tex-math notation="LaTeX">$\mathcal {H}$ </tex-math></inline-formula> by relaxing to <inline-formula> <tex-math notation="LaTeX">$(\varepsilon,\delta)$ </tex-math></inline-formula>-differential privacy. We apply our hypothesis selection algorithm to give learning algorithms for a number of natural distribution classes, including Gaussians, product distributions, sums of independent random variables, piecewise polynomials, and mixture classes. Our hypothesis selection procedure allows us to generically convert a cover for a class to a learning algorithm, complementing known learning lower bounds which are in terms of the size of the packing number of the class. As the covering and packing numbers are often closely related, for constant <inline-formula> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula>, our algorithms achieve the optimal sample complexity for many classes of interest. Finally, we describe an application to private distribution-free PAC learning.

Subfields: Artificial Intelligence, Computational Theory and Mathematics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

Citations in the corpus

Learning discrete distributions: user vs item-level privacy (2020, methodology): “Algorithm 1 Private hypothesis selection: PHS( H , D, α, ε ) [Bun et al., 2019] 1: Input : H = { H 1 , . . . , H d } the set of hypotheses, dataset D of s samples drawn i.i.d. from p ∈ H , accuracy parameter α ∈ (0 , 1), privacy parameter ε .”
Private Mean Estimation of Heavy-Tailed Distributions (2020, methodology): “In particular, in a similar way, we choose from the cover by setting up several pairwise comparisons, and privatize using the exponential mechanism in the same way as [BKSW19].”
CoinPress: Practical Private Mean and Covariance Estimation (2020, unknown): “(3)Under pure or concentrated DP, the dependence on R and K must be polylogarithmic [27, 5].”
Differentially Private Assouad, Fano, and Le Cam (2020, methodology): “Unless mentioned, the bounds are all for estimation under total variation distance. mean assumption is unavoidable, since by [BKSW19], it is not possible to learn a single Gaussian distribution under pure DP without this assumption.”