This chapter goes beyond the supervised learning of Chapter 7. It covers learning richer representation and learning what to do; this enables learning to be combined with reasoning. First we consider unsupervised learning in which the classifications are not given in the training set. This is a special case of learning belief network, which is considered next. Finally, we consider reinforcement learning, in which an agent learns how to act while interacting with an environment.

Clustering

Chapter 7 considered supervised learning, where the target features that must be predicted from input features are observed in the training data. In clustering or unsupervised learning, the target features are not given in the training examples. The aim is to construct a natural classification that can be used to cluster the data.

The general idea behind clustering is to partition the examples into clusters or classes. Each class predicts feature values for the examples in the class. Each clustering has a prediction error on the predictions. The best clustering is the one that minimizes the error.