CoreSetSelection<T>

Models & Types · AiDotNet.ActiveLearning

Implements core-set selection using the k-center-greedy algorithm for active learning.

Implements: IActiveLearningStrategy<T>

For Beginners

Core-set selection aims to select samples that best represent the overall data distribution. It uses the k-center algorithm which iteratively selects the sample that is farthest from all previously selected samples. This ensures good coverage of the feature space.

How It Works

Algorithm (k-center-greedy):

Advantages:

Complexity: O(n × k × d) where n=pool size, k=batch size, d=feature dimension.

Reference: Sener & Savarese, “Active Learning for Convolutional Neural Networks: A Core-Set Approach” (ICLR 2018).

Constructors

Constructor Summary
CoreSetSelection(DistanceMetric<T> distanceMetric) Initializes a new instance of the CoreSetSelection 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)