0
$\begingroup$

In the article about the convolution backpropagation, the computation of gradients to the input needs to rotate the weight and the computation of gradients to the weight also needs to rotate the input. However, in the article about the correlation backpropagation, only the computation of gradients to the input needs to rotate the weight. Why? Is this consistency due to the difference between convolution and correlation?

$\endgroup$

Your Answer

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

Browse other questions tagged or ask your own question.