tf_privacy.logistic_objective_perturbation

Trains and validates differentially private logistic regression model.

tf_privacy.logistic_objective_perturbation(
    train_dataset: datasets.RegressionDataset,
    test_dataset: datasets.RegressionDataset,
    epsilon: float,
    delta: float,
    epochs: int,
    num_classes: int,
    input_clipping_norm: float
) -> List[float]

The training is based on the Algorithm 1 of Kifer et al.

train_dataset consists of num_train many labeled examples, where the labels are in {0,1,...,num_classes-1}.
test_dataset consists of num_test many labeled examples, where the labels are in {0,1,...,num_classes-1}.
epsilon epsilon parameter in (epsilon, delta)-DP.
delta delta parameter in (epsilon, delta)-DP.
epochs number of training epochs.
num_classes number of classes.
input_clipping_norm l2-norm according to which input points are clipped.

List of test accuracies (one for each epoch) on test_dataset of model trained on train_dataset.