Skip to Main Content
Table 2: 
Accuracy of membership inference per classifier type, Perceptron (P), Decision Tree (DT), Naïve Bayes (NB), Nearest Neighbors (NN), and Multi-layer Perceptron (MLP). Alice column shows the accuracy of attack on Alice probes 𝒜in_probe and 𝒜out_probe. Bob columns show the accuracy on the classifiers’ train, validation, and test set. Note that, following the evaluation protocol explained in Section 4.3, only Carol the evaluator can observe the accuracy of the attacks on Alice model.
AliceBob:trainBob:validBob:test
50.0 50.0 50.0 50.0 
DT 50.4 51.4 51.2 51.1 
NB 50.4 51.2 51.1 51.0 
NN 49.9 61.6 50.5 50.0 
MLP 50.2 50.8 50.8 50.8 
AliceBob:trainBob:validBob:test
50.0 50.0 50.0 50.0 
DT 50.4 51.4 51.2 51.1 
NB 50.4 51.2 51.1 51.0 
NN 49.9 61.6 50.5 50.0 
MLP 50.2 50.8 50.8 50.8 
Close Modal

or Create an Account

Close Modal
Close Modal