0
$\begingroup$

I'm starting to learn how convolutional neural networks work, and I have a question regarding the filters. Are these chosen manually or are they generated by the network in training? If it's the latter, are the coefficients in the filters chosen at random, and then as the network is trained they are "corrected"?

Any help or insight you might be able to provide me in this matter is greatly appreciated!

$\endgroup$

1 Answer 1

1
$\begingroup$

The values in the filters are parameters that are learned by the network during training. When creating the network the values are initialized randomly according to some initialization scheme (e.g. Kaiming He initialization) and then during training are updated to achieve a lower loss (i.e. the learning process).

$\endgroup$
1
  • $\begingroup$ Very clear. Thank you so much! $\endgroup$ Commented May 15, 2022 at 9:54

Your Answer

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

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