Coalitional strategies for efficient individual prediction explanation

TL;DR

This paper introduces coalitional strategies for explaining individual ML predictions, which are more efficient than existing methods like SHAP, enabling faster and trustworthy explanations.

Contribution

It proposes coalitional methods for prediction explanation that improve computational efficiency while maintaining explanation accuracy, compared to traditional approaches.

Findings

Coalitional methods outperform SHAP in computation time.

Explanation accuracy remains acceptable with coalitional methods.

Methods facilitate wider practical application of ML explanations.

Abstract

As Machine Learning (ML) is now widely applied in many domains, in both research and industry, an understanding of what is happening inside the black box is becoming a growing demand, especially by non-experts of these models. Several approaches had thus been developed to provide clear insights of a model prediction for a particular observation but at the cost of long computation time or restrictive hypothesis that does not fully take into account interaction between attributes. This paper provides methods based on the detection of relevant groups of attributes -- named coalitions -- influencing a prediction and compares them with the literature. Our results show that these coalitional methods are more efficient than existing ones such as SHapley Additive exPlanation (SHAP). Computation time is shortened while preserving an acceptable accuracy of individual prediction explanations. Therefore, this enables wider practical use of explanation methods to increase trust between developed ML models, end-users, and whoever impacted by any decision where these models played a role.