Post-Hoc

Achieving Interpretability by applying Explainable AI methods to the model after it has been trained. They are often Model-Agnostic methods.

Can also be applied to intrinsically interpretable models.

• Permutation Feature Importance