We address the problem of non-parametric density estimation under the additional constraint that only privatised data are allowed to be published and available for inference. For this purpose, we adopt a recent generalisation of classical minimax theory to the framework of local $\alpha$-differential privacy and provide a lower bound on the rate of convergence over Besov spaces $B^s_{pq}$ under mean integrated $\mathbb L^r$-risk. This lower bound is deteriorated compared to the standard setup without privacy, and reveals a twofold elbow effect. In order to fulfil the privacy requirement, we suggest adding suitably scaled Laplace noise to empirical wavelet coefficients. Upper bounds within (at most) a logarithmic factor are derived under the assumption that $\alpha$ stays bounded as $n$ increases: A linear but non-adaptive wavelet estimator is shown to attain the lower bound whenever $p \geq r$ but provides a slower rate of convergence otherwise. An adaptive non-linear wavelet estimator with appropriately chosen smoothing parameters and thresholding is shown to attain the lower bound within a logarithmic factor for all cases.
[1]
Martin J. Wainwright,et al.
Minimax Optimal Procedures for Locally Private Estimation
,
2016,
ArXiv.
[2]
Martin J. Wainwright,et al.
Local Privacy and Minimax Bounds: Sharp Rates for Probability Estimation
,
2013,
NIPS.
[3]
D. Donoho,et al.
Geometrizing Rates of Convergence, III
,
1991
.
[4]
Pencho Petrushev,et al.
Needlet algorithms for estimation in inverse problems
,
2007,
0705.0274.
[5]
I. Johnstone,et al.
Density estimation by wavelet thresholding
,
1996
.
[6]
Larry A. Wasserman,et al.
Differential privacy for functions and functional data
,
2012,
J. Mach. Learn. Res..
[7]
A. Tsybakov,et al.
Wavelets, approximation, and statistical applications
,
1998
.
[8]
L. Wasserman,et al.
A Statistical Framework for Differential Privacy
,
2008,
0811.2501.