Corpus ID: 51935803

Test without Trust: Optimal Locally Private Distribution Testing

@inproceedings{Acharya2019TestWT,
  title={Test without Trust: Optimal Locally Private Distribution Testing},
  author={Jayadev Acharya and Cl{\'e}ment L. Canonne and Cody R. Freitag and Himanshu Tyagi},
  booktitle={AISTATS},
  year={2019}
}
We study the problem of distribution testing when the samples can only be accessed using a locally differentially private mechanism and focus on two representative testing questions of identity (goodness-of-fit) and independence testing for discrete distributions. First, we construct tests that use existing, general-purpose locally differentially private mechanisms such as the popular RAPPOR or the recently introduced Hadamard Response for collecting data and show that our proposed tests are… Expand

Tables and Topics from this paper

Differentially Private Testing of Identity and Closeness of Discrete Distributions
TLDR
The fundamental problems of identity testing (goodness of fit), and closeness testing (two sample test) of distributions over $k$ elements, under differential privacy are studied, and Le Cam's two point theorem is used to provide a general mechanism for proving lower bounds. Expand
Communication-Constrained Inference and the Role of Shared Randomness
TLDR
This work proposes a public-coin protocol that outperforms simulate-and-infer for distribution testing and is, in fact, sample-optimal for distribution learning and examines the role of shared randomness as a resource. Expand
Inference Under Information Constraints II: Communication Constraints and Shared Randomness
TLDR
This work proposes a general-purpose simulate-and-infer strategy that uses only private-coin communication protocols and is sample-optimal for distribution learning, and proposes a public-coin protocol that outperforms simulate- and-Infer for distribution testing and is, in fact, sample- optimal. Expand
The structure of optimal private tests for simple hypotheses
Hypothesis testing plays a central role in statistical inference, and is used in many settings where privacy concerns are paramount. This work answers a basic question about privately testing simpleExpand
Inference Under Information Constraints III: Local Privacy Constraints
TLDR
It is shown that the availability of shared (public) randomness greatly reduces the sample complexity and under the notion of local differential privacy, simple, sample-optimal, and communication-efficient protocols are proposed for these two questions in the noninteractive setting. Expand
Goodness-of-fit testing for H\"older continuous densities under local differential privacy
We address the problem of goodness-of-fit testing for Hölder continuous densities under local differential privacy constraints. We study minimax separation rates when only noninteractive privacyExpand
Locally private non-asymptotic testing of discrete distributions is faster using interactive mechanisms
TLDR
It is proved that general information theoretical bounds that allow us to establish the optimality of the authors' algorithms among all pairs of privacy mechanisms and test procedures, in most usual cases, are correct. Expand
Private Testing of Distributions via Sample Permutations
TLDR
The framework of property testing is used to design algorithms to test the properties of the distribution that the data is drawn from with respect to differential privacy, which indicates that differential privacy can be obtained in most regimes of parameters for free. Expand
Successive Refinement of Privacy
TLDR
This work provides (order-wise) tight characterizations of privacy-utility-randomness trade-offs in several cases for distribution estimation, including the standard LDP setting under a randomness constraint, and provides a non-trivial privacy mechanism for multi-level privacy. Expand
Private Two-Terminal Hypothesis Testing
TLDR
It is shown that, in general, meaningful correctness and privacy cannot be achieved if the users do not have access to correlated (but, not common) randomness, and the optimal correctness andPrivacy error exponents are characterized when the users has access to non-trivial correlated randomness. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 43 REFERENCES
Locally Private Hypothesis Testing
TLDR
This work discusses the randomized-response mechanism and shows that, in essence, it maps the null- and alternative-hypotheses onto new sets, an affine translation of the original sets, and gives sample complexity bounds for identity and independence testing under randomized- response. Expand
Differentially Private Testing of Identity and Closeness of Discrete Distributions
TLDR
The fundamental problems of identity testing (goodness of fit), and closeness testing (two sample test) of distributions over $k$ elements, under differential privacy are studied, and Le Cam's two point theorem is used to provide a general mechanism for proving lower bounds. Expand
Revisiting Differentially Private Hypothesis Tests for Categorical Data
TLDR
A modified equivalence between chi-squared tests and likelihood ratio tests is shown, more suited to hypothesis testing with privacy, and differentially private likelihood ratio and chi-Squared tests for a variety of applications on tabular data are developed. Expand
Local Private Hypothesis Testing: Chi-Square Tests
TLDR
This work analyzes locally private chi-square tests for goodness of fit and independence testing, which have been studied in the traditional, curator model for differential privacy, to explore the design of private hypothesis tests in the local model. Expand
Differentially Private Chi-Squared Hypothesis Testing: Goodness of Fit and Independence Testing
Hypothesis testing is a useful statistical tool in determining whether a given model should be rejected based on a sample from the population. Sample data may contain sensitive information aboutExpand
Differentially Private Identity and Closeness Testing of Discrete Distributions
TLDR
The theoretical results show that there exist private identity and closeness testers that are nearly as sample-efficient as their non-private counterparts, and an experimental evaluation of the algorithms on synthetic data shows that they achieve small type I and type II errors with sample size sublinear in the domain size of the underlying distributions. Expand
Minimax Optimal Procedures for Locally Private Estimation
TLDR
Private versions of classical information-theoretical bounds, in particular those due to Le Cam, Fano, and Assouad, are developed to allow for a precise characterization of statistical rates under local privacy constraints and the development of provably (minimax) optimal estimation procedures. Expand
RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response
TLDR
This paper describes and motivates RAPPOR, details its differential-privacy and utility guarantees, discusses its practical deployment and properties in the face of different attack models, and gives results of its application to both synthetic and real-world data. Expand
A New Approach for Testing Properties of Discrete Distributions
  • Ilias Diakonikolas, D. Kane
  • Computer Science, Mathematics
  • 2016 IEEE 57th Annual Symposium on Foundations of Computer Science (FOCS)
  • 2016
TLDR
The sample complexity of the algorithm depends on the structure of the unknown distributions - as opposed to merely their domain size - and is significantly better compared to the worst-case optimal L1-tester in many natural instances. Expand
Discrete Distribution Estimation under Local Privacy
TLDR
New mechanisms are presented, including hashed K-ary Randomized Response (KRR), that empirically meet or exceed the utility of existing mechanisms at all privacy levels and demonstrate the order-optimality of KRR and the existing RAPPOR mechanism at different privacy regimes. Expand
...
1
2
3
4
5
...