Decision Trees

Abstract

A decision tree is an algorithm for computing a function of an unknown input. Each node of the tree is labeled by a variable and the branches from that node are labeled by the possible values of the variable. The leaves are labeled by the output of the function. The process starts at the root, knowing nothing, works down the tree, choosing to learn the values of some of the variables based on those already known and eventually reaches a decision. The decision tree complexity of a function is the minimum depth of a decision tree that computes that function.