Logistic regression, Y \in {0, 1}?

[alanqrwang]

Hi, does the code support logistic regression models, i.e. Y \in {0, 1}? Is it possible to adapt the code to support this?

[hannahbus]

Hi, I am not sure if this is still relevant to you, but the methodology was indeed developed for binary classification tasks using a linear classifier. Although the underlying loss is defined as a 0-1 loss, the authors also conducted an analysis using logistic regressions, as explained in the section "Experiment" under the subheading "Ad hoc protocol." Disclaimer: I am not affiliated with the authors; I just happened to have read the paper and noticed your issue.