How can I give notations to the whole CNN network? Like, if I feed input 'x' to feature extractor, we will get the extracted feature embedding vector at the end that will be passed to the linear classifier and at the end we will apply the softmax? Can any one explain this whole process in the notations.

  • $\begingroup$ What have you tried so far? $\endgroup$ Commented Mar 22, 2022 at 20:26
  • $\begingroup$ X∈X,X is training set,the x is fed to the feature extractor we will get c= C(x;u)where u is weight tensor then this c will be passed to the linear classifier,g(c;w),w is weight tensor. the predictive probability after softmax function will be p(x)=σ(g(c;w)) $\endgroup$ Commented Mar 22, 2022 at 20:33
  • $\begingroup$ How would you do it for a different kind of neural network? CNNs are neural networks like any other neural network, as my drawing hopefully demonstrates. $\endgroup$
    – Dave
    Commented Mar 22, 2022 at 21:09
  • $\begingroup$ Yes, but I wanted professional notation for the CNN $\endgroup$ Commented Mar 23, 2022 at 7:36


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.