Can you please explain to me why here no answer is correct?
We are missing smoothness or a bounded gradient assumption here. That would be required to get a rate here.
Hope this helps.
Hi Thijs, if I understand correctly, according to the lecture notes, we have the tame strong case which has a rate of O(1/e), and it does not require smoothness or bounded grad.
But I think that the reason that no answer is correct in this case, is the fact that it is specifying the rate of the last iterate and not the average one (which I think would be O(1/e)). Am I missing anything?