I'm worrying that my network has become too complex. I don't want to end up with half of the network doing nothing but just take up space and resources.

So, what are the techniques for detecting and preventing overfitting to avoid such problems?

  1. Usually you keep track of training loss and validation loss and apply proper regularization technique (L1, L2, dropout, dropconnect, ...).

  2. The more interesting technique is to observe your validation loss with respect to the number of parameters in the network (often controlled by the number of layers/feature maps). If the validation starts dropping with raising your model's complexity, then your optimization might be bad or simply the model remembers all of the training samples and overfits badly.

| improve this answer | |

Your Answer

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

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