DensityWeightedSampling<T>
Models & Types · AiDotNet.ActiveLearning
Implements density-weighted sampling for active learning.
Implements: IActiveLearningStrategy<T>
For Beginners
Density-weighted sampling combines uncertainty with density weighting to avoid selecting outliers. Even if a sample is uncertain, it may not be informative if it’s an outlier that doesn’t represent the data distribution.
How It Works
Formula: Score(x) = Uncertainty(x) × Density(x)^β
where Density(x) is computed using average distance to k nearest neighbors.
Parameters:
Advantages:
Reference: Settles, B. & Craven, M. (2008). “An Analysis of Active Learning Strategies for Sequence Labeling Tasks.”
Constructors
| Constructor | Summary |
|---|---|
DensityWeightedSampling(double beta, int kNeighbors) |
Initializes a new instance of the DensityWeightedSampling class. |
Properties
| Property | Summary |
|---|---|
bool UseBatchDiversity { get; set; } |
|
string Name { get; } |
Methods
| Method | Summary |
|---|---|
Dictionary<string, T> GetSelectionStatistics() |
|
Vector<T> ComputeInformativenessScores(IFullModel<T, Tensor<T>, Tensor<T>> model, Tensor<T> unlabeledPool) |
|
int[] SelectSamples(IFullModel<T, Tensor<T>, Tensor<T>> model, Tensor<T> unlabeledPool, int batchSize) |