What is the linear regression?
Linear regression model refers to the relationship between the dependent and independent variables is linear. Similar to a linear regression:
Given a set of data collection ,
It functions as a linear combination of
Reference links: linear regression prediction method
What is the loss function?
Loss function: to measure the error between the predicted value and the true value. Generally use the mean square error
Which is solving the linear regression value, actual value
The overall mean square error
among them
What is the optimization function?
And a linear function of the above mentioned loss function is relatively simple, such predictions is called solving analytical solution (analytical solution). Involves a more complex level, such as the depth of learning, we generally limited by the optimization algorithm iterations model parameters to reduce the value of the loss function as possible. Such solution is called numerical solution (numerical solution).
Recommended Links: How to understand the stochastic gradient descent