Linear regression
Also: ordinary least squares, OLS, least squares regression
A model ŷ = w·x + b fitted by minimizing mean squared error, either in closed form (the normal equations) or with gradient descent. It is the simplest regression model and the base that many others extend.
ŷ = w·x + b, w* = (XᵀX)⁻¹ Xᵀy
Example
The points (1, 3), (2, 5) and (3, 7) are fit exactly by y = 2x + 1 with MSE 0, so the model predicts 9 for x = 4.