Privacy-preserving statistical estimation with optimal convergence rates

By Adam D. Smith. 2011

348 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Networks and Communications, 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 Artificial Intelligence

In Computer Networks and Communications

In Statistics and Probability

In Privacy & Security

Citations in the corpus

The structure of optimal private tests for simple hypotheses (2018, background): “[20, 32, 45, 69]) and others where there is a significant cost, even for relaxed definitions of privacy (e.”
Differential Privacy Preservation for Deep Auto-Encoders: an Application of Human Behavior Prediction (2016, methodology): “There are several potential solutions for -differential privacy that target regression, which is similar to the data reconstruction and cross-entropy error functions in deep autoencoders (Zhang et al. 2012; Chaudhuri and Monteleoni 2008; Chaudhuri, Monteleoni, and Sarwate 2011; Lei 2011; Smith 2011)”
Crypto-Nets: Neural Networks over Encrypted Data (2014, background): “Another line of work focuses on differential privacy (Chaudhuri et al., 2011; Duchi et al., 2012; Dwork, 2008; Smith, 2011; Wasserman & Zhou, 2010).”
Functional Mechanism: Regression Analysis under Differential Privacy (2012, methodology): “Smith [27] proposes a general framework for statistical analysis that utilizes both the Laplace mechanism and exponential mechanism.”
Minimax Optimal Procedures for Locally Private Estimation (2016, background): “…work aims to link differential privacy to statistical concerns (Dwork and Lei 2009; Wasserman and Zhou 2010; Hall, Rinaldo, and Wasserman 2012; Smith 2011; Chaudhuri, Monteleoni, and Sarwate 2011; Rubinstein et al. 2012); in particular, researchers have developed algorithms for private robust…”
Private selection from private candidates (2018, background): “The problem of private median finding, and more generally private percentile estimation has been studied in several works [5, 10, 30, 33].”
Learning with Differential Privacy: Stability, Learnability and the Sufficiency and Necessity of ERM Principle (2015, background): “…testing, and hence fits naturally in the general framework of statistical machine learning, e.g., see (Dwork & Lei, 2009; Wasserman & Zhou, 2010; Smith, 2011; Lei, 2011; Wang et al., 2015), as well as applications involving regression (Chaudhuri et al., 2011; Thakurta & Smith, 2013) and GWAS…”
Convergence Rates for Differentially Private Statistical Estimation (2012, background): “In further work, Smith (2011) shows how to construct a diﬀerentially private approximation A T to certain types of statistical estimators T , and establishes asymptotic normality of his estimator provided certain conditions on T hold.”
Differential privacy for functions and functional data (2012, background): “In all these cases, the data (both the input and output) are assumed to be real numbers or vectors.”
Adversarial machine learning (2019, methodology): “Numerous other practical learning algorithms have been made differentially private including regularized logistic regression [10], several collaborative filtering algorithms [43], point estimation [59], nearest neighbor, histograms, perceptron [6], and more.”
Finite Sample Differentially Private Confidence Intervals (2017, background): “ the entire real line). Thus, there have been several works seeking the best asymptotic approximations we can obtain for population statistics under diﬀerential privacy, such as Dwork and Lei (2009); Smith (2011); Wasserman and Zhou (2010); Wasserman (2012); Hall et al. (2013); Duchi et al. (2013a,b”
Private Mean Estimation of Heavy-Tailed Distributions (2020, background): “Smith [Smi11] gives an algorithm which allows one to estimate asymptotically normal statistics with optimal convergence rates, but no ﬁnite sample complexity guarantees.”
Local Privacy, Data Processing Inequalities, and Statistical Minimax Rates (2013, background): “There are, however, results that are more favorable for differential privacy; for example, Smith [48] shows that in some parametric problems, the non-local form of differential privacy [24] can be satisfied while yielding asymptotically optimal parametric rates of convergence for some point estimato”
Bounding User Contributions: A Bias-Variance Trade-off in Differential Privacy (2019, background): “Another line of work similar to the one introduced here is that of Smith (2011).”
A New Class of Private Chi-Square Tests (2016, background): “Smith [20] then showed that for a large family of statistics, there is a corresponding differentially private statistic that shares the same asymptotic distribution as the original statistic.”
Differentially Private K-Means Clustering (2015, methodology): “[25, 31] propose the sample and aggregate framework (SAF), and use k-means clustering as a motivating application for SAF.”