See the article on my own blog https://dyingdown.github.io/2019/09/03/Model-representation/
To describe supervised learning,we establish notation for future use.
** x i x^i xi **denote the input variables also called input features
y i y^i yi denote the output variables also called target variables that we are trying to predict.
A pair **( x i , y i x^i, y^i xi,yi) **is called a training set.
Note: i is just a superscript implying an index to the training set, and has nothing to do with exponentiation.
X denote the space of input values.
Y denote the space of output values.
[外链图片转存失败(img-pN3FdLLD-1568186465444)(https://d3c33hcgiwev3.cloudfront.net/imageAssetProxy.v1/H6qTdZmYEeaagxL7xdFKxA_2f0f671110e8f7446bb2b5b2f75a8874_Screenshot-2016-10-23-20.14.58.png?expiry=1567641600000&hmac=I6c3DPvjbzUTdQXfZdN2BtRjiy98uNxta1mcly2eP4A)]
If the target variables are continuous, we call the learning problem a regression problem.
If Y can take on only a small number of discrete values, we call the learning problem a regression problem.