Ensemble Learning combines multiple base models (e.g., decision trees or neural networks) to produce a single, more robust prediction. By leveraging the diversity of individual models, ensembles often outperform standalone algorithms. Techniques include bagging, boosting, and stacking.