ExpectedModelChange<T>

Models & Types · AiDotNet.ActiveLearning

Implements Expected Model Change (EMC) for active learning.

Implements: IActiveLearningStrategy<T>

For Beginners

Expected Model Change selects samples that would cause the largest change to the model’s parameters if they were labeled and used for training. The intuition is that samples which significantly change the model provide the most learning value.

How It Works

How it works:

Variants:

Reference: Settles et al., “Multiple-Instance Active Learning” (2008). ICML.

Constructors

Constructor Summary
ExpectedModelChange(ChangeMetric<T> metric) Initializes a new instance of the ExpectedModelChange 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)