Interpretable AutoML

Surrogate Models

Simple and interpretable models trained to approximate the behavior of a complex model, allowing generation of global or local explanations of the original model's decisions.

← Wstecz