GaussianDifferentialPrivacyVector<T>

Models & Types · AiDotNet.FederatedLearning.Privacy

Implements Gaussian differential privacy for vector-based model updates.

Inheritance: FederatedLearningComponentBase<T> → PrivacyMechanismBase<Vector<T>, T> → GaussianDifferentialPrivacyVector<T>

How It Works

For Beginners: This adds carefully calibrated random noise to a parameter vector so that individual data points cannot be inferred from the update, while the overall signal remains useful.

Constructors

Constructor Summary
GaussianDifferentialPrivacyVector(double clipNorm, int? randomSeed)

Methods

Method Summary
Vector<T> ApplyPrivacy(Vector<T> model, double epsilon, double delta)
double GetClipNorm() Gets the gradient clipping norm used for sensitivity bounding.
double GetPrivacyBudgetConsumed()
string GetMechanismName()
void ResetPrivacyBudget() Resets the privacy budget counter.