0
$\begingroup$

I don't understand the difference between the parameter $C$ and $λ$ in terms of the SVM. It seems to me that they are both involved in regulating over-fitting of the data.

What difference between $C$ and $λ$?

$\endgroup$
  • 1
    $\begingroup$ Are you familiar with the primal and dual statements of the SVM optimization problem? $\endgroup$ – Matthew Drury Aug 19 '17 at 19:39
  • 1
    $\begingroup$ @MatthewDrury I'm not familiar with the primal and dual statements of the SVM optimization problem. $\endgroup$ – ComputerScienceLearner Aug 19 '17 at 22:30
1
$\begingroup$

Both are regularisation hyperparameters, related by $C \sim \frac{1}{\lambda}$. Which one you use will depend on the formulation of SVM that you're using. As mentioned on one of your question comments, it would be good to read into primal and dual statements for SVMs if you're feeling a bit lost - Chapter 12 in Elements of Statistical Learning covers this.

$\endgroup$

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.