When I am reading about convolutional neural networks, I have encountered the following sentence from the textbook(page 341) that says about the limitation of the usage of the convolution in CNNs.

When a task involves incorporating information from very distant locations in the input, then the prior imposed by convolution may be inappropriate

My interpretations are

  1. when the object is very small, then the convolution may not work well.

  2. when the object is very large, then the components of the object are far away from each other and hence the convolution may not work well.

Which of the interpretation is correct? If both are wrong, then what is the correct interpretation.

If possible, please provide an example to understand it.

  • 1
    $\begingroup$ I think that your second interpretation is correct because the kernel is typically a lot smaller than the input. Maybe this answer may provide some intuition. $\endgroup$
    – nbro
    May 15 at 11:27

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.