Locally Differentially Private Protocols for Frequency Estimation

By Tianhao Wang, Jeremiah Blocki, Ninghui Li, Somesh Jha. 2017

650 citations recorded by the source corpus.

Subfields: Statistics and Probability, 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 Statistics and Probability

In Privacy & Security

Citations in the corpus

Breaking the Communication-Privacy-Accuracy Trilemma (2020, background): “Frequency estimation under local differential privacy has been studied in [35], where they propose schemes for estimating the frequency of an individual symbol and minimizing the variance of the estimator.”
Federated Heavy Hitters Discovery with Differential Privacy (2019, background): “…of fundamental work shows that in the context of learning distributions and heavy hitters, local DP often leads to a signiﬁcant reduction in utility (Kairouz et al., 2014; Wang et al., 2017; Bassily et al., 2017; Kairouz et al., 2016; Ye and Barg, 2018; Duchi et al., 2013; Cormode et al., 2018).”
Amplification by Shuffling: From Local to Central Differential Privacy via Anonymity (2018, background): “Due to their attractive trust model, LDP mechanisms have recently received signiﬁcant industrial adoption for the privacy-preserving collection of heavy hitters [EPK14, App17, DKY17], as well as increased academic attention [BS15, BNST17, QYY + 16, WBLJ17].”
Marginal Release Under Local Differential Privacy (2017, methodology): “[41], although we find that these make little difference compared to the “vanilla” Randomized Response probabilities.”
Prochlo: Strong Privacy for Analytics in the Crowd (2017, background): “Local differential privacy is the approach taken by our RAPPOR software monitoring system for the Chrome Web browser [28].”
BLENDER: Enabling Local Search with a Hybrid Differential Privacy Model (2017, background): “Very recently, much attention has been given to the heavy hitters problem in the local differential privacy setting both from a theoretical and an applied perspective [38, 9, 24].”
Collecting and Analyzing Multidimensional Data with Local Differential Privacy (2019, methodology): “, [5], [18], [38]) for categorical data are designed for this purpose, albeit limited to a single categorical attribute.”
PROCHLO: Strong Privacy for Analytics in the Crowd (2017, background): “Another major obstacle to the practical use of locallydifferentially-private methods—based on our experiences with RAPPOR—is the opaque, fixed, and statistical nature of the data collected.”
Local Differential Private Data Aggregation for Discrete Distribution Estimation (2019, background): “[39], [40] also induced k-subset mechanism in the context of categorical data distribution estimation, but lacks characterization on its mutual information bound/channel capacity.”
Distributed Clustering in the Anonymized Space with Local Differential Privacy (2019, background): “Based on differential privacy, the notion of LDP (Local Differential Privacy) is also proposed to protect local privacy context from data analysis [16], [17], [18], [19].”
AttriGuard: A Practical Defense Against Attribute Inference Attacks via Adversarial Machine Learning (2018, methodology): “Local differential privacy (LDP): LDP [33–40] is a technique based on ε-differential privacy [41] to protect privacy of an individual user’s data record, i.”
Calibrate: Frequency Estimation and Heavy Hitter Identification with Local Differential Privacy via Incorporating Prior Knowledge (2018, methodology): “Local differential privacy (LDP) [27], [5], [20], [22], [30], [3], [12], [13], [34], [29], [23], [8], [24], [14], [6], [28], [32], [35], [1], a privacy protection mechanism based on differential privacy [15], can address both challenges.”
CALM: Consistent Adaptive Local Marginal for Marginal Release under Local Differential Privacy (2018, methodology): “But in the local setting, several previous work pointed out that by partitioning users into groups, and having each group use the full privacy budget, the overall error will be smaller [26, 37, 38].”
Practical and Robust Privacy Amplification with Multi-Party Differential Privacy (2019, methodology): “By applying Theorem 3 together with Theorem 2 from [46], we can see the following.”
Locally Differentially Private Heavy Hitter Identification (2017, background): “pare different mechanisms using estimation variance [35].”
Locally Differentially Private Data Collection and Analysis (2019, methodology): “In this context, from Theorem 2 of [48], the variance for the number of times that the true value occurs among N users’ noisy values will be Var = Nq(1− q) (p− q)2 + Nfv(1 − p− q) p− q , (43) where fv is the frequency of the value v ∈ [1, k].”