0
$\begingroup$

I just have a question and I did not find clear explanation, hinge loss is upper bound of 0-1 loss, so does this mean upper bound regard to a single point loss such as 0 or 1 ?!. So, for example the 0-1 loss for set of points is 0 or 1, and for hinge loss we have continuous loss values from 0 to inf. So, by comparing the set of loss values for 0-1 loss and hinge loss we find the lower bound is 0 and upper bound is 1. I am right?!.

This is the link: https://www.di.ens.fr/~fbach/learning_theory_class/lecture3.pdf. Please see the page 3, the first paragraph under the title "Conditional Φ-risk and classification calibration".

Thanks

Improve this question
$\endgroup$
2
  • $\begingroup$ Could you give us more details and some context on where did you find the statement you are mentioning? $\endgroup$
    – Tim
    Commented Jun 11, 2021 at 17:50
  • $\begingroup$ This is the link: di.ens.fr/~fbach/learning_theory_class/lecture3.pdf. Please see the page 3, the first paragraph under the title "Conditional Φ-risk and classification calibration". $\endgroup$
    – miss Ran
    Commented Jun 11, 2021 at 18:19

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.