Expand description
Various implementations of Measures (and associated Distance).
A Measure is used to measure the distance between distributions. The distance is expressed in terms of an associated type.
Example
MaxDivergence<Q> has an associated distance type of Q.
This means that the symmetric distance between vectors is expressed in terms of the type Q.
In this context Q is usually f32 or f64.
Structs
$(\epsilon, \delta)$-approximate differential privacy.
$\epsilon$-pure differential privacy.
A function mapping from $\delta$ to $\epsilon$.
$\epsilon(\delta)$-approximate differential privacy.
$\rho$-zero concentrated differential privacy.