IActiveLearningStrategy<T>
Interfaces · AiDotNet.Interfaces
Defines a strategy for active learning that selects the most informative samples for labeling from a pool of unlabeled data.
Known implementations: BALD<T>, BatchBALD<T>, CoreSetSelection<T>, DensityWeightedSampling<T>, DiversitySampling<T>, EntropySampling<T>, ExpectedModelChange<T>, HybridSampling<T>, InformationDensity<T>, LeastConfidenceSampling<T>, MarginSampling<T>, QueryByCommittee<T>, RandomSampling<T>, UncertaintySampling<T>, VariationRatios<T>
For Beginners
Active learning helps when labeling data is expensive or time-consuming. Instead of randomly selecting samples to label, active learning intelligently picks the samples that would be most helpful for training the model. This can dramatically reduce the number of labels needed while achieving similar or better performance.
How It Works
Common strategies include:
Typical Usage Flow:
Properties
| Property | Summary |
|---|---|
bool UseBatchDiversity { get; set; } |
Gets or sets whether to use batch-mode selection that considers diversity among selected samples. |
string Name { get; } |
Gets the name of this active learning strategy. |
Methods
| Method | Summary |
|---|---|
Dictionary<string, T> GetSelectionStatistics() |
Gets statistics about the most recent sample selection. |
Vector<T> ComputeInformativenessScores(IFullModel<T, Tensor<T>, Tensor<T>> model, Tensor<T> unlabeledPool) |
Computes informativeness scores for all samples in the unlabeled pool. |
int[] SelectSamples(IFullModel<T, Tensor<T>, Tensor<T>> model, Tensor<T> unlabeledPool, int batchSize) |
Selects the most informative samples from the unlabeled pool for labeling. |