Stanford MachineLearning 第一周学习笔记

What is machine learning:

什么是机器学习：

1.the field of study that gives computers the ability to learn without being explicitly programmed.

在特定编程的情况下，就能让计算机有学习能力的一个研究领域

2.A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.

 

Any machine learning problem can be assigned to following methods:

      SupervisedLearning(监督学习):

      A already know correct output data set and already know relationship between the input and the output

      UnsupervisedLearning(非监督学习)

     Unsupervised learning allows us to approach problems with little or no idea what our results should look like. We can derive structure from data where we don't necessarily know the effect of the variables

 

Model representation

一般用x(i)变量代表input， y(i)代表output。为了更正式的描述监督学习，还要加入一个训练集（training set）这个概念，以及h（hypothesis）假设

通常对于我们需要预测的是一个连续性的问题，我们一般称之为regression problem（回归问题），而当y是数量值或者很少数的值的时候，我们称之为classification problem（分类问题）

 

Cost function

我们可以通过代价函数来衡量预测的准确性。