On A Step by Step Backpropagation Example, in the section The Backwards Pass - Output Layer, after the graphic, the author wrote:
$\frac{\partial E_{total}}{\partial out_{o1}}=2* \frac{1}{2}(target_{o1}-out_{o1})^{2-1}*-1+0$
Please explain where the factor $-1$ comes from.
The $-1$ factor comes from the chain rule. When differentiating $$(x - y)^2$$ partially with respect to $y$, you differentiate the outer square function, leaving the inner function in tact, and then multiply by the derivative of the inner function $$\frac{\partial}{\partial y} (x - y) = -1.$$
