22. Overview of Gaussian processes

In this chapter we introduce and demonstrate Gaussian processes (GPs).

Good references are Gaussian processes for machine learning by Rasmussen and William; The Kernel Cookbook by Duvenaud; Chapter 21 of Bayesian Data Analysis, 3rd Edition by Gelman et al.; and Melendez et al., Phys. Rev. C 100, 044001 (2019), arXiv:1904.10581.

The sections here are:

• Intuition for Gaussian process from simulations, which points to several websites with Gaussian process visualizations and tasks to build intuition;

• Basic info on GPs, which places GPs in the context of stochastic processes and provides a first exposure to the mathematical form of GPs and its use for interpolation and regression;

• More on Gaussian processes, which distinguishes parametric and non-parametric inference

• Scikit-learn demo notebooks uses demonstrations with a popular library;