A model has a single weight www, and its loss is
L(w)=2w2−6w+5L(w) = 2w^2 - 6w + 5L(w)=2w2−6w+5
Training starts at w0=4w_0 = 4w0=4 and takes one gradient descent step with learning rate η=0.1\eta = 0.1η=0.1, using the update rule
w1=w0−η L′(w0)w_1 = w_0 - \eta \, L'(w_0)w1=w0−ηL′(w0)
What is w1w_1w1?
Select all that apply.
555
333
2.72.72.7
−6-6−6