The monitor as a noisy classifier: ROC & AUC
Model the trusted monitor's suspicion scores as two overlapping distributions. Drag its quality to see the ROC curve, AUC, and how many backdoors slip past a fixed 2% false-positive threshold.
Innocent solutions Backdoored solutions
Suspicion-score distributions
ROC curve
AUC
0.877
Backdoors caught @ 2% FPR
34.0%
Missed (false negatives)
66.0%