Machine Learning

Machine Learning

we say that a machine learns with respect to a particular task T,performance   metric P,and type of experience E,if the system reliably improves its performance P at task T,following experience E

我们说，如果系统能够可靠地提供其在任务T中的性能P和经验E，那么机器就能从特定的任务T、性能指标 P和经验 E中学习。

Supervised learning    监督式学习

Classification  predicting  class  labels

Regression  predicting continuous outcomes

Reinforcement learning

Solving interactive problem with Reinforcement learning

 

 

Unsupervised Learning

Discover hidden structures with unsupervised Learning

Performance measure 

性能估计

 Classification  loss

Accuracy

Regression loss

（平均绝对错误）