Class FederatedHeterogeneityCorrectionOptions

Namespace

AiDotNet.Models.Options

Assembly

AiDotNet.dll

Configuration options for federated heterogeneity correction algorithms.

public class FederatedHeterogeneityCorrectionOptions

Inheritance

object

FederatedHeterogeneityCorrectionOptions

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

For Beginners: In federated learning, clients often have different data and different compute speeds. Heterogeneity correction methods help reduce "client drift" so the global model converges more reliably on non-IID data.

Properties

Algorithm

Gets or sets the heterogeneity correction algorithm.

public FederatedHeterogeneityCorrection Algorithm { get; set; }

Property Value

FederatedHeterogeneityCorrection

ClientLearningRate

Gets or sets the client learning rate used by methods that need it (e.g., SCAFFOLD control variates).

public double ClientLearningRate { get; set; }

Property Value

double

FedDynAlpha

Gets or sets the FedDyn regularization strength (alpha).

public double FedDynAlpha { get; set; }

Property Value

double