quantileDataAccess {DPpack}R Documentation

Differentially Private Quantile Data Access Function

Description

This function performs the data access step in the computation of a differentially private quantile. The utility vector is computed as in Smith (2011), while the sensitivities are calculated based on bounded and unbounded differential privacy (Kifer and Machanavajjhala 2011) according to the theoretical values (Gillenwater et al. 2021).

Usage

quantileDataAccess(x, quant, lower.bound, upper.bound)

Arguments

x

Numeric vector.

quant

Real number between 0 and 1 indicating which quantile to return.

lower.bound

Real number giving the lower bound of the input data.

upper.bound

Real number giving the upper bound of the input data.

Value

List of a vector corresponding to the utility function, the sorted and clipped vector of inputs and the sensitivity calculated based on theoretical values.

References

Kifer D, Machanavajjhala A (2011). “No Free Lunch in Data Privacy.” In Proceedings of the 2011 ACM SIGMOD International Conference on Management of Data, SIGMOD '11, 193–204. ISBN 9781450306614, doi:10.1145/1989323.1989345.

Smith A (2011). “Privacy-Preserving Statistical Estimation with Optimal Convergence Rates.” In Proceedings of the Forty-Third Annual ACM Symposium on Theory of Computing, STOC '11, 813–822. ISBN 9781450306911, doi:10.1145/1993636.1993743.

Gillenwater J, Joseph M, Kulesza A (2021). “Differentially Private Quantiles.” In Meila M, Zhang T (eds.), Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proceedings of Machine Learning Research, 3713–3722. http://proceedings.mlr.press/v139/gillenwater21a/gillenwater21a.pdf.

Examples

quantileDataAccess(c(1,1,-2,8,-6),.25,-10,10)


[Package DPpack version 0.2.2 Index]