Bayesian Analysis with Python

Chapter 8. Gaussian Processes

All models that we have seen so far were parametric models. These are models with a fixed number of parameters that we are interested in estimating. Another type of models are those known as non-parametric models. Non-parametric models are models where the number of parameters increases with the data, in other words, models with a potentially infinite number of parameters that we somehow manage to reduce to a finite number, just those necessary to describe the data. We will began the chapter, by learning about the concept of a kernel, and how to rethink problems in terms of kernels. Gaussians are the workhorse of statistics and this is not only true for classical methods, but also Bayesian statistics and machine learning. ...