add_gaussian_noise

Adds Gaussian (normal) noise to a numeric value or vector for (epsilon, delta)-differential
privacy. The Gaussian mechanism provides (epsilon, delta)-DP and is often used when
delta &gt; 0 is acceptable.

internal

A tidy-style interface for applying differential privacy to data frames.
Provides pipe-friendly functions to add calibrated noise, compute private statistics,
and track privacy budgets using the epsilon-delta differential privacy framework.
Implements the Laplace mechanism (Dwork et al. 2006 <doi:10.1007/11681878_14>) and
the Gaussian mechanism for achieving differential privacy as described in Dwork and
Roth (2014) <doi:10.1561/0400000042>.

Thomas Tarler

tidydp

Tidy Differential Privacy

add_gaussian_noise function

<dl><dt>x</dt>
<dd>Numeric value or vector to add noise to</dd>
<dt>sensitivity</dt>
<dd>The L2 sensitivity of the query</dd>
<dt>epsilon</dt>
<dd>Privacy parameter (smaller = more privacy)</dd>
<dt>delta</dt>
<dd>Privacy parameter (probability of privacy breach), typically very small</dd></dl>

Arguments

Add Gaussian Noise — add_gaussian_noise

<dl>

<dt>x</dt>
<dd>Numeric value or vector to add noise to</dd>


<dt>sensitivity</dt>
<dd>The L2 sensitivity of the query</dd>


<dt>epsilon</dt>
<dd>Privacy parameter (smaller = more privacy)</dd>


<dt>delta</dt>
<dd>Privacy parameter (probability of privacy breach), typically very small</dd>

</dl>

add_gaussian_noise: Add Gaussian Noise

Description

Usage

Value

Arguments