Bagging (bootstrap aggregating)
Ensemble technique that trains multiple models on bootstrap samples of the training data and aggregates their predictions to reduce variance.
Bagging, short for bootstrap aggregating, is an ensemble method in machine learning that trains many models on different random subsamples of the same dataset. Each subsample is drawn with replacement from the training set, so some examples appear multiple times and others are left out. Predictions from the individual models are then combined, typically by majority vote for classification or averaging for regression, which reduces variance compared with a single model fit to the full data.
Leo Breiman introduced bagging in 1996, and it became a building block of random forests, where many decision trees trained via bagging are further decorrelated by random feature selection at each split. Bagging remains a standard tool in classical supervised learning pipelines for tabular data. See the random forests article and timeline entry.