RecurrentHyperNetAlgorithm<T, TInput, TOutput>
Models & Types · AiDotNet.MetaLearning.Algorithms
Implementation of Recurrent HyperNetwork for meta-learning.
Inheritance: ModelBase<T, TInput, TOutput> → MetaLearnerBase<T, TInput, TOutput> → RecurrentHyperNetAlgorithm<T, TInput, TOutput>
Implements: IGeneratedParameterRegistrar<T>
How It Works
A GRU-like recurrent cell processes compressed gradient information at each adaptation step, maintaining hidden state that captures the optimization trajectory. The recurrent output is used to compute per-parameter learning rate modulation factors, enabling adaptive step sizes that evolve through the inner loop based on gradient history.
Algorithm:
Constructors
| Constructor | Summary |
|---|---|
RecurrentHyperNetAlgorithm(RecurrentHyperNetOptions<T, TInput, TOutput> options) |
Properties
| Property | Summary |
|---|---|
MetaLearningAlgorithmType AlgorithmType { get; } |
Methods
| Method | Summary |
|---|---|
IModel<TInput, TOutput, ModelMetadata<T>> Adapt(IMetaLearningTask<T, TInput, TOutput> task) |
|
T MetaTrain(TaskBatch<T, TInput, TOutput> taskBatch) |