1
$\begingroup$

I would like to ask about SVM. I want to compare between 0-1 loss and hinge loss functions. My question is how to compare between them?!. Should we construct different SVM models, which each for particular loss function?! or just construct one SVM and evaluate the SVM model by 0-1 loss and hinge loss. I confused whether loss function should include in the process of construction SVM model or not.

Thank you for any help you can offer.

$\endgroup$
1
$\begingroup$

Without choosing a loss function, you cannot construct the SVM because you won't have an optimization objective to work on. So, you first choose your loss function and construct your model. In the end, there will be two different models to compare.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Thanks for help, do you know whether hard margin and soft margin are related to 0-1 loss and hinge loss respectively, or can be part of both loss functions. $\endgroup$ – miss Ran Jan 18 at 20:09
  • $\begingroup$ In hard-margin svm, there maybe no solution at all, and you're only interested in the margin part, i.e. minimizing $||w||$. Since the cost of misclassification, $C$, is infinity in hard-margin svm, there is no difference if you use hinge or something else. $\endgroup$ – gunes Jan 19 at 10:16

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.