Feature Importance versus Feature Influence and What It Signifies for Explainable AI

TL;DR

This paper clarifies the distinction between feature importance and influence in explainable AI, introduces the CIU method for unified global and local explanations, and demonstrates its advantages over influence-only approaches.

Contribution

The paper introduces the Contextual Importance and Utility (CIU) method, providing a unified framework for global and local feature explanations in post-hoc XAI.

Findings

CIU offers more expressive explanations than influence-only methods.

CIU demonstrates high fidelity and stability in explanations.

The method applies to both global and local interpretability tasks.

Abstract

When used in the context of decision theory, feature importance expresses how much changing the value of a feature can change the model outcome (or the utility of the outcome), compared to other features. Feature importance should not be confused with the feature influence used by most state-of-the-art post-hoc Explainable AI methods. Contrary to feature importance, feature influence is measured against a reference level or baseline. The Contextual Importance and Utility (CIU) method provides a unified definition of global and local feature importance that is applicable also for post-hoc explanations, where the value utility concept provides instance-level assessment of how favorable or not a feature value is for the outcome. The paper shows how CIU can be applied to both global and local explainability, assesses the fidelity and stability of different methods, and shows how explanations that use contextual importance and contextual utility can provide more expressive and flexible explanations than when using influence only.