TrimmedMeanFullModelAggregationStrategy<T, TInput, TOutput>
Models & Types · AiDotNet.FederatedLearning.Aggregators
Coordinate-wise trimmed mean aggregation for IFullModel.
Inheritance: FederatedLearningComponentBase<T> → AggregationStrategyBase<IFullModel<T, TInput, TOutput>, T> → RobustFullModelAggregationStrategyBase<T, TInput, TOutput> → TrimmedMeanFullModelAggregationStrategy<T, TInput, TOutput>
How It Works
For Beginners: This strategy sorts each parameter across clients, drops the extreme values on both ends, then averages the remaining values. This reduces the impact of outliers.
Constructors
| Constructor | Summary |
|---|---|
TrimmedMeanFullModelAggregationStrategy(double trimFraction) |
Methods
| Method | Summary |
|---|---|
IFullModel<T, TInput, TOutput> Aggregate(Dictionary<int, IFullModel<T, TInput, TOutput>> clientModels, Dictionary<int, double> clientWeights) |
|
string GetStrategyName() |