Techniques that combine multiple learning models to improve performance. Examples include bagging (bootstrap aggregating) and boosting. Ensembles tend to yield more accurate and robust predictions than single models by averaging over different hypotheses. In forex applications, ensemble methods (like random forests or gradient-boosted trees) help trading systems generalize better from noisy market data. They reduce the risk of overfitting to peculiarities of one dataset by blending many models, making trading signals more reliable.

Back to Robots Glossary Index