Differentially Private Learning of Structured Discrete Distributions

By Ilias Diakonikolas, Moritz Hardt, Ludwig Schmidt. 2015

64 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods, Privacy & Security

How researchers see this paper

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

In Probabilistic & Bayesian Methods

In Privacy & Security

Citations in the corpus

Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning (2017, background): “Several works have adopted DP as an efficient defense mechanism such as [5, 7, 11, 12, 18, 23, 24, 37, 53, 60, 65, 72, 86].”
Learning discrete distributions: user vs item-level privacy (2020, methodology): “…with local diﬀerential privacy [Duchi et al., 2013, Kairouz et al., 2016, Acharya et al., 2019a, Ye and Barg, 2018], with global diﬀerential privacy [Diakonikolas et al., 2015, Acharya et al., 2020], and with communication constraints [Barnes et al., 2019, Acharya et al., 2019a], among others.”
Federated Heavy Hitters Discovery with Differential Privacy (2019, methodology): “estimation both in the central and local models of DP (Diakonikolas et al., 2015; Bonomi & Xiong, 2013; Xu et al., 2016; Zhou & Lin, 2018; Wang et al., 2017; Acharya et al., 2018; Bassily et al., 2017; Kairouz et al., 2016; Ye & Barg, 2018; Avent et al., 2017; Bun et al., 2018).”
Discrete Distribution Estimation under Local Privacy (2016, background): “e a trusted service provider releases randomized data (e.g., NIH releasing medical records) to protect sensitive user information (Dwork, 2006; Dwork et al., 2006; Dwork & Lei, 2009; Dwork, 2008; Diakonikolas et al., 2015; Blocki et al., 2016). 2. Preliminaries 2.1. Local differential privacy Le”
Priv'IT: Private and Sample Efficient Identity Testing (2017, background): “The problem of distribution estimation (rather than testing) has also recently been studied under the lens of differential privacy (Diakonikolas et al., 2015).”
Differentially Private Identity and Equivalence Testing of Discrete Distributions (2018, background): “…testers (matched by information-theoretic lower bounds) (Paninski, 2008; Daskalakis et al., 2013; Chan et al., 2014; Valiant & Valiant, 2014; Diakonikolas et al., 2015b;c; Acharya et al., 2015; Diakonikolas & Kane, 2016; Canonne et al., 2016; Diakonikolas et al., 2016; Canonne et al.,…”
Differentially Private Identity and Closeness Testing of Discrete Distributions (2017, result): “Finally, recent work of Diakonikolas et al. [12] has provided differentially private algorithms for learning various families of discrete distributions.”
Private Hypothesis Selection (2019, background): “Private learning algorithms for the latter class, piecewise polynomials, have been studied by Diakonikolas, Hardt, and Schmidt [DHS15].”
Membership Inference Attacks Against Machine Learning Models (2016, background): “Differential privacy [12] has been applied to linear and logistic regression [8], [37], support vector machines [28], risk minimization [5], [9], [34], deep learning [1], [30], learning an unknown probability distribution over a discrete population from random samples [10], and releasing hyper-param”
Mutual Information Optimally Local Private Discrete Distribution Estimation (2016, background): “…privacy guarantee, and achieves plenty of theoretical results or mechanisms characterizing the deﬁnition, both in centralized database setting (e.g. in [21][4][13]) and in the local setting [18] (e.g. in [6][14][16]), where each data provider sanitizes their secret data locally and independently.”
INSPECTRE: Privately Estimating the Unseen (2018, background): “There has also been study on private distribution learning (Diakonikolas et al., 2015; Duchi et al., 2017; Karwa & Vadhan, 2018; Acharya et al., 2018; Kamath et al., 2018), in which we wish to estimate parameters of the distribution, rather than just a particular property of interest.”
Privately Learning High-Dimensional Distributions (2018, background): “These are standard concentration inequalities, and a reference for these facts is Section 4 of Diakonikolas et al. (2016). Fact 25 Let X1, .”
Differentially Private Algorithms for Learning Mixtures of Separated Gaussians (2019, methodology): “With the above theorem in our tookit, we can finally show that the each estimated Gaussian is close to its respective actual Gaussian to within α in TV-distance.”
Differentially Private Testing of Identity and Closeness of Discrete Distributions (2017, background): “[24] studies structured distribution estimation under differential privacy.”
Differentially Private Assouad, Fano, and Le Cam (2020, background): “(Diakonikolas et al. (2015), Theorem 13) Ω ( k α2 + k α(ε+δ) )”
Private Mean Estimation of Heavy-Tailed Distributions (2020, unknown): “similar to techniques in Diakonikolas et al. (2017) (see Lemma A.”
Local Differential Private Data Aggregation for Discrete Distribution Estimation (2019, background): “, in [27], [28]), the participants’ data are collected and managed by a trustable database curator, who then release sanitized statistics of the database to protect individual’s data privacy.”