Wraps the given class in GradientInversionServerWrapper. Returns: cls: GradientInversionServerWrapper

Wraps the given class in DistributedBackdoorAttackClientWrapper. Returns: cls: a class wrapped in DistributedBackdoorAttackClient

Wraps the given class in ModelReplacementAttackClientWrapper. Returns: cls: a class wrapped in ModelReplacementAttackClientWrappe

Attaches DPSGD to the client class. Args: cls: Client class. Returns: DPSGDClientWrapper: Wrapped c

Attack victim model Args: x: target datasets which the attacker wants to classify y: target labels which the attacker

Generates fake images Args: n (int): the number of fake images created by the Generator Returns:

Performs the DIVA white-box attack on input data. Args: data (tuple): A tuple containing input data and corresponding labels.

Performs the FGSM attack on input seed data. Args: data (tuple): A tuple containing input seed data and corresponding labels.

Executes evasion attack Args: x0 (np.array) : initial data point Returns: xm (np.array) : created adversaria

Calculates leak_auc on the given SplitNN model reference: https://arxiv.org/abs/2102.08504 Args: dataloader (t

Create an adversarial example for poison attack Args: xc: initial attack point yc: true label of initial attack point

Execute backward mode automatic differentiation with the give loss. Args: loss (torch.Tensor): the value of calculated loss.

Balance adjustment during deletion in the left subtree

Balance adjustment during deletion in the left subtree

Balance adjustment during deletion in the right subtree

Balance adjustment during deletion in the right subtree

* Iterators */

* Bucket interface */

* Bucket interface */

* Return the bucket [0, bucket_count()) to which the hash belongs. * If bucket_count() is 0, it must always return 0. */

Compute log(A_alpha) for any positive finite alpha.

Certify the robustness of the model. Args: counts (torch.Tensor): Count of predictions. Returns: fl

* Reset the growth policy as if it was created with a bucket count of 0. * After a clear, the policy must always return 0 when bucket_for_hash is

uint64 maxTxId = 2;

uint64 minTxId = 1;

float tof = 6;

int32 toi = 4;

string tos = 5;

.storage.TupleData.Type type = 3;

Computes the cosine similarity distance between fake and received gradients. Args: fake_gradients (list of torch.Tensor): List of fake gr

* Lookup */

Delete the node with key 'key' from the tree rooted at 'this'

Delete the minimum key from the subtree rooted at 'this' Returchildren the new minimum key in the subtree rooted at 'this'