Słownik AI
Kompletny słownik sztucznej inteligencji
Feature Randomness
Introduction of randomness in the selection of predictor variables for each decision tree, reducing correlation between ensemble models.
Bagging Estimator
Composite model resulting from the application of bagging, combining predictions of multiple base estimators by voting or averaging.
Bootstrap Sample Size
Number of observations in each bootstrap sample, typically equal to the size of the original training set to maximize diversity.
Parallel Learning
Ability of bagging to train base models independently and simultaneously, offering significant computational advantages.
Model Diversity
Measure of dissimilarity between predictions of base models, essential for bagging effectiveness and obtained through bootstrap sampling.
Bootstrap Distribution
Empirical distribution of statistics calculated on multiple bootstrap samples, used to estimate prediction uncertainty.
Bagging Classifier
Specific implementation of bagging for classification problems, typically using majority voting to combine predictions.
Bagging Regressor
Version of bagging adapted for regression problems, combining predictions by averaging or median of base models.