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？

exists some ‘underlying pattern’ to be learned
- so 'performance' measure can be improved;

but no (easy) programmable definition
- so 'ML' is needed

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: $x\in X$
output: $y \in Y$
unknown pattern to be learned $\Leftrightarrow$ target function:
$f: X \rightarrow Y$
data: training examples : $D = {(x_1,y_1),(x_2,y_2),\dots,(x_N,y_N)}$
hypothesis $\Leftrightarrow$ skill hopefully good performance:
$g = X \rightarrow Y$ ('learned' formula to be used)

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

Machine Learning and Other Fields

Machine Learning and Data Mining

Machine learning use data to compute hypothesis $g$ that approximates target function $f$
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 $g$ that approximates target function $f$
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 $g$ that approximates target function $f$
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。