Ensemble Methods in Machine Learning.¶
Dietterich, T. G. (2000). Ensemble Methods in Machine Learning. Multiple Classifier Systems.
Cited by¶
3 citations across 3 artifacts.
Each citation links to the sentence it supports in the citing article.
Primes¶
- Correlation
- In every case the operative fact is the correlation among the components, not their individual behavior — the lower the correlation, the greater the variance reduction from pooling.
This sourceShows that pooling weakly correlated predictors lowers aggregate prediction variance — the same correlation-driven variance-reduction arithmetic underlying portfolio diversification and redundancy design.
- In every case the operative fact is the correlation among the components, not their individual behavior — the lower the correlation, the greater the variance reduction from pooling.
- Minority Signal Preservation
- Ensemble methods — bagging, boosting, and random forests preserve weak learners whose individual signals are dominated; the aggregate extracts value from the diversity.
This sourceShows that retaining and aggregating diverse weak learners (bagging, boosting, error-correcting output coding) outperforms selecting the single best predictor.
- Ensemble methods — bagging, boosting, and random forests preserve weak learners whose individual signals are dominated; the aggregate extracts value from the diversity.
- Population Coding
- Ensemble methods — bagging, boosting, random forests, and mixtures of experts produce predictions from a population of weak learners whose joint accuracy exceeds the best individual.
This sourceSurveys bagging, boosting, and error-correcting output coding as populations of classifiers whose joint accuracy exceeds the best individual.
- Ensemble methods — bagging, boosting, random forests, and mixtures of experts produce predictions from a population of weak learners whose joint accuracy exceeds the best individual.
Verification¶
This reference passed the adversarial substantiation pipeline: it was checked to exist and to support the claim it is attached to. See how references were verified.
Registry ID ref:a7465a5f1dee · see in the full table