机器学习:P1-P2 Introduction

Machine Learning ~ Looking for a Function From Data

Step 1 : A set of function (Model)

Step 2 : Goodness of function f

Step 3 : pick the best function

Learning Map

 

Regression :

The output of the target function f is "scalar"

Classification :

• Binary Classification :such as spam filtering

• Multi-class Classification: such as Document Classification

Supervised Learning :

Hard to collect a large amount of labelled data

Semi-supervised Learning ：

Labelled data && Unlabeled data

Transfer Learning:

Labelled data && data not related to the task considered

Unsupervised Learning:

machine reading,creating new things,Machine Drawing

structured learning:

Speech recognition , Machine Translation , face recognition ......

Reinforcement Learning:

learning from critics ， Alpha Go : supervised +reinforcement