Model Explanations via the Axiomatic Causal Lens

TL;DR

This paper introduces axiomatic causal explanation measures for black-box models, integrating causal responsibility and game-theoretic influence, and provides a formal framework connecting explanations with causality and influence measures.

Contribution

It proposes three novel axiomatic explanation measures based on causal responsibility and influence, bridging model explanations with causal and game-theoretic concepts.

Findings

The measures can be uniquely characterized by axioms.

The approach extends to compute Shapley-Shubik and Banzhaf indices.

Empirical analysis on the Adult-Income dataset demonstrates their effectiveness.

Abstract

Explaining the decisions of black-box models is a central theme in the study of trustworthy ML. Numerous measures have been proposed in the literature; however, none of them take an axiomatic approach to causal explainability. In this work, we propose three explanation measures which aggregate the set of all but-for causes -- a necessary and sufficient explanation -- into feature importance weights. Our first measure is a natural adaptation of Chockler and Halpern's notion of causal responsibility, whereas the other two correspond to existing game-theoretic influence measures. We present an axiomatic treatment for our proposed indices, showing that they can be uniquely characterized by a set of desirable properties. We also extend our approach to derive a new method to compute the Shapley-Shubik and Banzhaf indices for black-box model explanations. Finally, we analyze and compare the necessity and sufficiency of all our proposed explanation measures in practice using the Adult-Income dataset. Thus, our work is the first to formally bridge the gap between model explanations, game-theoretic influence, and causal analysis.