Lazy Versus Eager Learning

• Radial Basis Function is eager, all the other IBLers are lazy
• lazy methods use less time in training and more during prediction
• lazy methods may consider the query instance when deciding how to generalize
• lazy learners have the advantage of a richer hypothesis space because it uses many different local functions to form its implicit global approximation
• eager learners, even RBFs, cannot customize itself to unknown future query instances