In the previous chapter, we discussed various forms of non-Machine Learning based Procedural Content Generation, but before we dive into Procedural Content Generation via Machine Learning (PCGML), we must first introduce Machine Learning (ML). At its core, ML is a set of techniques for learning functions—a process for finding a mapping from input to output. In the case of PCGML we try to learn a function that will procedurally generate content (i.e., map from some input space to some generative space). Note that “learn” is a very overloaded term, as machine learning doesn’t learn like a person, but instead determines a function that best fits some data according to a set of assumptions about the nature of the data. In this chapter, we discuss the basics of machine learning—defining a problem domain, determining the form of function we want to learn, learning the parameters of that function, and assessing the results. In the following chapter we will discuss practical considerations, but for now let’s walk through a simple example.