amad-person
commented
3 years ago Hi @tr0p1x, thanks for opening the issue.
From what I can see, in fact, the error comes from the way SKLearn sets the thresholds for computing the false positive rate and the true positive rate when passing an array with float values.
Yes, you're right. sklearn is computing the threshold after seeing the data instead of fixing it at 0.5. However, this is desired behaviour for our library - because in membership inference attacks we want to find the best threshold that separates out members and non-members.
Hope this clears things up!
luigitropiano
In some situations there seems to be problems with how the ROC curve is calculated. This is clearly visible when there is a very high number of False Positives or False Negatives. I'll post an example where IMHO the auc should be 0.5 (and not 1.0), as all my samples are classified as positives and there are no negatives.
The error is probably due to following line in meminf.py: https://github.com/privacytrustlab/ml_privacy_meter/blob/abf42232dfd712bc377db690190dfdff53db7eac/ml_privacy_meter/attack/meminf.py#L592
I think it should be something like:
fpr, tpr, _ = roc_curve(target, probs > 0.5)From what I can see, in fact, the error comes from the way SKLearn sets the thresholds for computing the false positive rate and the true positive rate when passing an array with float values: If your target vector is
[[1], [0], [1], [1], [0]]and your predictions vector (probs) is[[0.9], [0.9], [0.9], [0.9], [0.9]]the auc(fpr, tpr) value returned by SKLearn is 0.5, (meaning there are 0 false negatives and 2 false positives, and that's fine), but if your predictions vector is[[0.9], [0.8], [0.9], [0.9], [0.8]]the auc(fpr, tpr) value returned by SKLearn is 1, (like there were 0 false negatives and 0 false positives))If I get this right, this is an unwanted behavior for your use-case, as you want an auc of 0.5 in both cases (both 0.8 and 0.9 are values greater then 0.5 and a prediction value greater than 0.5 should count as false positive if the target value is 0)
By doing
probs > 0.5, however, it returns a vector of booleans of the same size as probs with True where the original values were greater then 0.5 and with False otherwhere; passing this vector instead of probs seems to work.