机器学习-03-上.pdf

机器学习第三讲 （上） 有指导学习的一般原理 卢志茂 我们的问题  什么是基于实例的学习  什么是假设类 hypothesis class  什么是归纳偏倚什么是归纳偏倚 inductive biasinductive bias  什么是泛化 generalization  什么是欠拟合 underfitting  什么是过拟合 overfitting  基于实例的学习  学习类别：家用汽车  属性为：价格和发动机的马力  描述方法：  假设类  Equation 2.4 fixes H, the hypothesis class from which we believe C is drawn, namely, the set of rectangles.  The learning algorithm then finds hypothesis the particular hypothesis, h ∈ H, to approximate C as closely as possible.   我们学习的目的 ：  经验误差 emprirical error  假设类的能力  泛化generalization  the hypothesis class H is the set of all possible rectangles. Each 1h 2h h h quadruple (p , p , e , e ) defines one hypothesis, h, from H, 1 2 and we need to choose the best one, or in other words, we need to find the values of these four parameters given the training set, to include all the positive examples and none of the negative examples.  最特殊的假设最特殊的假设 most special hypothesis  最一般假设 most general hypothesis  训练集相容  零误差的假设情况 h ∈H between S and G is a valid hypothesis with no error, said to be consistent with the training set  通常在给定的假设X 、H 条件下，会有多个S 、 G i j  如果X 足够大，可以存在唯一的S G 、  For our error function to have a minimum at h with the maximum margin, we should use an use an error (loss) function error (loss) function which not only checks whether an instance is on the correct side of the boundary but also how far away it is.  许多领域 ，S与G之间的实例