ABSTRACT
In the previous chapters, we have introduced many of the main concepts that underpin machine learning methods. We have seen how, for a particular model, we can choose parameters and make predictions based on observed data. This has been done in three ways – finding the parameters that minimise a loss function, finding those that maximise a likelihood function and by treating the parameters as random variables. We will meet some of these approaches again in this and subsequent chapters as we tackle the main algorithmic families that make up the field of machine learning: classification, clustering and projection.
