Lecture 1：The Learning Problem

What is machine learning ？

learning：acquiring skill with accumulated experience for observations
machine learning: acquiring skill with accumulated/computed from data

而skill便是提升某些性能指标。一个典型应用就是ML通过股票数据分析预测如何更好的赚钱。
那什么时候用到机器学习呢？

1. exists some ‘underlying pattern’ to be learned
    - so 'performance' measure can be improved;
2. but no (easy) programmable definition
    - so 'ML' is needed
3. there is data from the pattern
    - so ML has some 'inputs' learn from

Components of Learning

以下是基本的记号：

• input:
• output:
• unknown pattern to be learned target function:
    
• data: training examples :
• hypothesis skill hopefully good performance:
     ('learned' formula to be used)

而machine learning要做的事情就是：use data to compute hypothesis that approximates target function

Machine Learning and Other Fields

Machine Learning and Data Mining

Machine learning use data to compute hypothesis that approximates target function
Data Mining use huge data to find property that is interesting

但是，其实在现实当中区别机器学习和数据挖掘并不是一件简单的事情，毕竟二者之间相辅相成，在传统意义上，数据挖掘也是致力于在大数据集上实现高效的计算。

Machine Learning and Artificial Intelligence

Machine learning use data to compute hypothesis that approximates target function
Artificial Intelligence compute something that shows intelligence behavior
也就是说，机器学习是人工智能实现的一种途径。

Machine Learning and Statistics

Machine learning use data to compute hypothesis that approximates target function
Statistics use data to inference about an unknown process
传统的统计学习的关注点在于证明数学假设，对于计算方面涉猎不足，然而统计学习为ML提供了很多很好的工具。