ActiveLearningStrategyType
Enums · AiDotNet.Models.Options
Specifies the active learning strategy to use for sample selection.
For Beginners
Active learning strategies help identify which unlabeled samples would be most valuable to label. Different strategies use different criteria to measure how “informative” a sample is for training the model.
Fields
| Field | Summary |
|---|---|
ActiveLearningStrategyType BALD |
Bayesian Active Learning by Disagreement - uses mutual information between predictions and model parameters for selection. |
ActiveLearningStrategyType BatchBALD |
Batch-mode BALD that accounts for redundancy when selecting multiple samples. |
ActiveLearningStrategyType CoreSetSelection |
Selects representative samples that form a core set of the data. |
ActiveLearningStrategyType DensityWeightedSampling |
Weights uncertainty by sample density in the input space. |
ActiveLearningStrategyType DiversitySampling |
Selects samples that maximize coverage of the input space. |
ActiveLearningStrategyType EntropySampling |
Selects samples based on entropy of predicted class probabilities. |
ActiveLearningStrategyType ExpectedModelChange |
Selects samples that would cause the largest change to model parameters. |
ActiveLearningStrategyType HybridSampling |
Combines multiple criteria (uncertainty + diversity) for selection. |
ActiveLearningStrategyType InformationDensity |
Combines uncertainty with representativeness based on local density. |
ActiveLearningStrategyType LeastConfidenceSampling |
Selects samples where the top prediction has the lowest confidence. |
ActiveLearningStrategyType MarginSampling |
Selects samples with the smallest margin between top two predictions. |
ActiveLearningStrategyType QueryByCommittee |
Uses multiple models and selects samples where they disagree the most. |
ActiveLearningStrategyType Random |
Selects samples randomly. |
ActiveLearningStrategyType UncertaintySampling |
Selects samples where the model is most uncertain about predictions. |
ActiveLearningStrategyType VariationRatios |
Uses variation ratios (1 - max probability) for uncertainty estimation. |
int value__ |