Gaussian: Radial Kernels

In XiaoqiLu/PhD-Thesis: Regularized Q-Learning

Description

Utility functions for radial kernels.

Usage

Gaussian(r, scale = 1)

Multiquadric(r, scale = 1)

InverseQuadratic(r, scale = 1)

InverseMultiquadric(r, scale = 1)

Bump(r, scale = 1)

Arguments

r	a positive numeric object
scale	a positive number for scaling

Value

an object of the same size of r

Functions

• Gaussian: \exp(- r^2)

• Multiquadric: √{1 + r^2}

• InverseQuadratic: 1 / (1 + r^2)

• InverseMultiquadric: 1 / √{1 + r^2}

• Bump: \exp(- 1 / (1 - r^2))) when r < 1, and 0 otherwise. Note that it is not scaled (Bump(0) < 1) and it is compactly supported.

Examples

Gaussian(0 : 6, scale = 2)

XiaoqiLu/PhD-Thesis documentation built on March 1, 2021, 10:49 a.m.