A linear model y^=wx+b\hat{y} = wx + by^=wx+b is fitted to three training examples:
using mean squared error
L(w,b)=13∑i=13(wxi+b−yi)2L(w, b) = \frac{1}{3}\sum_{i=1}^{3}\left(wx_i + b - y_i\right)^2L(w,b)=31∑i=13(wxi+b−yi)2
Find the parameters that minimize LLL, together with the value of LLL there.
Select all that apply.
w=2.5w = 2.5w=2.5, b=0.5b = 0.5b=0.5, minimum L=1.5L = 1.5L=1.5
w=2.5w = 2.5w=2.5, b=0.5b = 0.5b=0.5, minimum L=0.5L = 0.5L=0.5
w=2.8w = 2.8w=2.8, b=0b = 0b=0, minimum L=0.6L = 0.6L=0.6
w=4w = 4w=4, b=−2b = -2b=−2, minimum L=0L = 0L=0