基于实例的学习课件.pptVIP

第8章 基于实例的学习 Instance Based Learning Instance Based Learning k-Nearest Neighbor Locally weighted regression Radial basis functions Case-based reasoning Lazy and eager learning Nearest neighbor Key idea: just store all training examples Nearest neighbor: Given query instance xq , first locate nearest training example xn , then estimate k - Nearest neighbor k - Nearest neighbor: Given xq , If discrete-valued target function: take vote among its k nearest nbrs If real-valued target function: take mean of f values of k- nearest nbrs When To Consider Nearest Neighbor Instances map to po