Class FederatedServerOptimizerOptions

Namespace

AiDotNet.Models.Options

Assembly

AiDotNet.dll

Configuration options for server-side federated optimizers (FedOpt family).

public class FederatedServerOptimizerOptions

Inheritance

object

FederatedServerOptimizerOptions

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

For Beginners: These optimizers run on the server after aggregation to update the global model. If Optimizer is None, the server uses the aggregated parameters directly (FedAvg-style).

Properties

Beta1

Gets or sets the beta1 coefficient (Adam/Yogi).

public double Beta1 { get; set; }

Property Value

double

Beta2

Gets or sets the beta2 coefficient (Adam/Yogi).

public double Beta2 { get; set; }

Property Value

double

Epsilon

Gets or sets the epsilon value used for numerical stability.

public double Epsilon { get; set; }

Property Value

double

LearningRate

Gets or sets the server learning rate (step size).

public double LearningRate { get; set; }

Property Value

double

Momentum

Gets or sets the server momentum coefficient for FedAvgM.

public double Momentum { get; set; }

Property Value

double

Optimizer

Gets or sets the server optimizer.

public FederatedServerOptimizer Optimizer { get; set; }

Property Value

FederatedServerOptimizer