Even-if Explanations: Formal Foundations, Priorities and Complexity

TL;DR

This paper investigates the computational complexity of semifactual explanations in explainable AI, compares interpretability across model types, and introduces a user-preference framework to personalize explanations, enhancing interpretability.

Contribution

It provides a formal analysis of semifactual explanations, compares model interpretability, and proposes a preference-based framework for personalized, user-centric explanations.

Findings

Linear and tree-based models are more interpretable than neural networks.

Complexity results for interpretability problems vary across models.

Algorithms are provided for polynomial cases in the framework.

Abstract

EXplainable AI has received significant attention in recent years. Machine learning models often operate as black boxes, lacking explainability and transparency while supporting decision-making processes. Local post-hoc explainability queries attempt to answer why individual inputs are classified in a certain way by a given model. While there has been important work on counterfactual explanations, less attention has been devoted to semifactual ones. In this paper, we focus on local post-hoc explainability queries within the semifactual `even-if' thinking and their computational complexity among different classes of models, and show that both linear and tree-based models are strictly more interpretable than neural networks. After this, we introduce a preference-based framework that enables users to personalize explanations based on their preferences, both in the case of semifactuals and counterfactuals, enhancing interpretability and user-centricity. Finally, we explore the complexity of several interpretability problems in the proposed preference-based framework and provide algorithms for polynomial cases.