Ensemble Learning

Ensemble Bootstrapping

A sampling method with replacement used to create varied training datasets, enabling the training of multiple models on different subsets to capture prediction variability. Particularly effective for uncertainty estimation in Model-Based RL.

← Quay lại