Equation 1.3
I understand from the equation 1.3 that on every iteration we are improving the line that will act as a classifier. This line is defined by the weights and we are improving the weights. So we take the last known weights W(t) and add y(t)X(t) to it to get the new and better weights W(t+1)
What I don't understand is how did we compute that the increase in W(t) to get W(t+1) should be y(t)X(t) ?
Thanks
