ActiveTransFSLAlgorithm<T, TInput, TOutput>
Models & Types · AiDotNet.MetaLearning.Algorithms
Implementation of ActiveTransFSL: Active Transductive Few-Shot Learning.
Inheritance: ModelBase<T, TInput, TOutput> → MetaLearnerBase<T, TInput, TOutput> → ActiveTransFSLAlgorithm<T, TInput, TOutput>
How It Works
ActiveTransFSL performs standard inductive adaptation on support data, then applies transductive refinement using query gradients. The refinement is focused on the most uncertain parameter dimensions (measured by gradient magnitude), implementing an active learning strategy in parameter space for transductive inference.
Algorithm:
Constructors
| Constructor | Summary |
|---|---|
ActiveTransFSLAlgorithm(ActiveTransFSLOptions<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) |