9
$\begingroup$

Especially in the computer-science oriented side of the machine learning literature, AUC (area under the receiver operator characteristic curve) is a popular criterion for evaluating classifiers. What are the justifications for using the AUC? E.g. is there a particular loss function for which the optimal decision is the classifier with the best AUC?

$\endgroup$
  • 1
    $\begingroup$ AUC is a loss function, it is clear that for this loss function the optimal decision is the classifier with the best AUC. $\endgroup$ – robin girard Jun 5 '11 at 19:23
  • 1
    $\begingroup$ @robingirard No it's not, since it's not differentiable, i.e. you can't optimize it directly. $\endgroup$ – cpury Jun 22 '18 at 15:20
13
$\begingroup$

For binary classifiers $C$ used for ranking (i.e. for each example $e$ we have $C(e)$ in the interval $[0, 1]$) from which the AUC is measured the AUC is equivalent to the probability that $C(e_1) > C(e_0)$ where $e_1$ is a true positive example and $e_0$ is a true negative example. Thus, choosing a model with the maximal AUC minimizes the probability that $C(e_0) \geq C(e_1)$. That is, minimizes the loss of ranking a true negative at least as large as a true positive.

$\endgroup$

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.