Shapley-Lorenz Decompositions in eXplainable Artificial Intelligence

Abstract

Explainability of artificial intelligence models has become a crucial issue, especially in the most regulated fields, such as health and finance. In this paper, we provide a global explainable AI model which is based on Lorenz decompositions, thus extending previous contributions based on variance decompositions. This allows the resulting Shapley Lorenz decomposition to be more generally applicable, and provides a unifying variable importance criteria that combines predictive accuracy with explainability, using a normalised and easy to interpret metric. The proposed decomposition is illustrated within the context of a real financial problem: the prediction of bitcoin prices.

Keywords: Shapley values, Lorenz Zonoids, Predictive accuracy

JEL Classification: C45, C53, C58, G32