ABSTRACT
This chapter looks at two of the common learning paradigms used in artificial intelligence (AI), both of which are also well known in statistics. These methods share an approach to learning that is based on exploiting regularities among observations, so that predictions are made on the basis of similar previously encountered situations. The methods differ, however, in the way that similarity is expressed; trees make important shared properties explicit, whereas instance-based approaches equate (dis)similarity with some measure of distance.
