Measuring Model Understandability by means of Shapley Additive Explanations

In this work we link the understandability of machine learning models to the complexity of their SHapley Additive exPlanations (SHAP). Thanks to this reframing we introduce two novel metrics for understandability: SHAP Length and SHAP Interaction Length. These are model-agnostic, efficient, intuitive and theoretically grounded metrics that are anchored in well-established game-theoretic and psychological principles. We show how these metrics resonate with other model-specific ones and how they can enable a fairer comparison of epistemically different models in the context of Explainable Artificial Intelligence. In particular, we quantitatively explore the understandability-performance trade-off of different models which are applied to both classification and regression problems. Reported results suggest the value of the new metrics in the context of automated machine learning and multi-objective optimisation.

Palabras clave: Explainable Artificial Intelligence, Explainable AI, Evaluation