Stability theory of game-theoretic group feature explanations for machine learning models

TL;DR

This paper develops a stability theory for game-theoretic feature explanations in machine learning, analyzing how different explanation methods behave under various conditions and proposing grouping strategies to enhance explanation stability.

Contribution

It introduces a stability framework for explanation operators based on game values and proposes feature grouping methods that stabilize explanations and unify different explanation types.

Findings

Marginal explanations can be discontinuous on certain domains.

Conditional explanations remain stable across various conditions.

Grouping features stabilizes explanations and unifies marginal and conditional methods.

Abstract

In this article, we study feature attributions of Machine Learning (ML) models originating from linear game values and coalitional values defined as operators on appropriate functional spaces. The main focus is on random games based on the conditional and marginal expectations. The first part of our work formulates a stability theory for these explanation operators by establishing certain bounds for both marginal and conditional explanations. The differences between the two games are then elucidated, such as showing that the marginal explanations can become discontinuous on some naturally-designed domains, while the conditional explanations remain stable. In the second part of our work, group explanation methodologies are devised based on game values with coalition structure, where the features are grouped based on dependencies. We show analytically that grouping features this way has a stabilizing effect on the marginal operator on both group and individual levels, and allows for the unification of marginal and conditional explanations. Our results are verified in a number of numerical experiments where an information-theoretic measure of dependence is used for grouping.