Open ahy69195 opened 2 years ago
I understand now, we take the partial derivatives (gradients) to find the new values of a0 and a1 by using this formula. Correct me if I'm wrong.
where b is the new value and a is the previous value and gamma is the learning rate (hyper parameter) and "( Δf(a) ) is simply the direction of the steepest descent."
How does the partial derivative become a0 = a0 - alpha (etc.) and a1 = a1 - alpha (etc.) ?