Now the question is how we can obtain the optimal w such that 
is minimized. We do so via gradient descent.
Gradient descent (also called steepest descent) is a procedure of minimizing an objective function by first-order iterative optimization. In each iteration, it moves a step that is proportional to the negative derivative of the objective function at the current point. This means the to-be-optimal point iteratively moves downhill towards the minimal value of the objective function. The proportion we just mentioned is called learning rate, or step size. It can be summarized in a ...
