Stacking: A technique for ensemble learning that involves training multiple models and combining their predictions by training a meta-model on the output of the base models.