0
$\begingroup$

I am trying to understand the loss function of autoencoders.

I think the loss function of the decoder is to compare the input image with the decoded image, tell me if I am correct.

but I don't understand what is being compared to find the encoder error.

Someone explains to me that it compares to find the encoder error. and decoder.

$\endgroup$
0

1 Answer 1

1
$\begingroup$

Yes, the loss of a normal autoencoder is simply the difference between the input image and the decoded image.

While encoder and decoder have different names, they are effectively part of the same neural network, so the prediction error is backpropagated through the decoder up to the encoder.

In more sophisticated autoencoders, like variational autoencoders, the loss may present extra terms, e.g. to enforce some structure in the output of the encoder.

$\endgroup$

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.