phi.gaussian {DPpack}R Documentation

Transform Function for Gaussian Kernel Approximation

Description

This function maps an input data row x with a given prefilter to an output value in such a way as to approximate the Gaussian kernel (Chaudhuri et al. 2011).

Usage

phi.gaussian(x, theta)

Arguments

x

Vector or matrix corresponding to one row of the dataset X.

theta

Randomly sampled prefilter vector of length n+1, where n is the length of x.

Value

Mapped value corresponding to one element of the transformed space.

References

Chaudhuri K, Monteleoni C, Sarwate AD (2011). “Differentially Private Empirical Risk Minimization.” Journal of Machine Learning Research, 12(29), 1069-1109. https://jmlr.org/papers/v12/chaudhuri11a.html.

Examples

  x <- c(1,2,3)
  theta <- c(0.1, 1.1, -0.8, 3)
  phi.gaussian(x, theta)


[Package DPpack version 0.2.2 Index]