I am trying to learn how to do deep neural networks with this Ipython notebook. I'm puzzled about notations in linear backward learning section.

For layer $l$, the linear part is: $Z^{[l]} = W^{[l]} A^{[l-1]} + b^{[l]}$ (followed by an activation). I want to get $(dW^{[l]}, db^{[l]} dA^{[l-1]})$.

The course suppose that I have already calculated the derivative $dZ^{[l]} = \frac{\partial \mathcal{L} }{\partial Z^{[l]}}$. Yet, What does $\mathcal{L}$ stands for in the derivative $dZ^{[l]} = \frac{\partial \mathcal{L} }{\partial Z^{[l]}}$ means ? I have never seen it defined here.

  • 1
    $\begingroup$ Your links are behind the wall, one needs an account to see it $\endgroup$ – Aksakal Jul 30 '18 at 18:08
  • $\begingroup$ @Aksakal, sure. I've updated. $\endgroup$ – IggyPass Jul 31 '18 at 8:55

It probably means "loss" as in "loss function" but as with everything else in mathematical notation, this is merely a convention and anyone can define any symbol to mean anything. The only way to be sure is to ask the author.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.