Andrew Ng 's machine learning lecture note (1)

    Any machine learning can be thought as 3 parts . 

(1) Experience  ，E means the machine learn from what experience (In chinese means 需要怎样的学习过程)

(2)P,P means the performance measure. (In chinese means 测量这个结果的好坏 )

(3)Task,T  just means the task . ( In chinese means 需要完成的目标 )

 

Machine learning can be divided into 2 parts .  

(1)supervised learning

supervised learning means we have the data set {xi yi}(i=1~n)

supervised learning includes regression and classification . 

regression means we must get the actually input output function which is continual .

classification means  we must get the actually input output function which is discret . 

(2)unsupervised learning 

unsupervised learning means we have the data set {xi }(i=1~n)

unsupervised learning includes the cluster problem.

cluster problem means the computer get the data set then it automatically divides the data into several parts . 

Hypothesis

Hypothesis is a function that the computer learns from the data set . Hypothesis is a model.

Cost function . 

Cost function is used to calculate the difference between the  Hypothesis and the actual model . I would take an example .

error square function which is a kind of cost function . 

m : the number of the data set 

: the hypothesis

:one of the element of the data set  .

3/26

My mistake ， this function  is called square error cost function