noise constructor

Make a Measurement that adds noise from the appropriate distribution to the input.

Usage

make_noise(input_domain, input_metric, output_measure, scale, k = NULL)

Arguments

input_domain

Domain of the data type to be privatized.

input_metric

Metric of the data type to be privatized.

output_measure

Privacy measure. Either MaxDivergence or ZeroConcentratedDivergence.

scale

Noise scale parameter.

k

The noise granularity in terms of 2^k.

Value

Measurement

Details

Valid inputs for input_domain and input_metric are:

input_domain	input type	input_metric
atom_domain(T)	T	absolute_distance(QI)
vector_domain(atom_domain(T))	Vec<T>	l2_distance(QI)

Required features: contrib

make_noise in Rust documentation.

Supporting Elements:

• Input Domain: DI

• Output Type: MI

• Input Metric: MO

• Output Measure: DI::Carrier