Flexible and Robust Counterfactual Explanations with Minimal Satisfiable Perturbations

TL;DR

This paper introduces CEMSP, a framework for generating minimal, robust counterfactual explanations by constraining feature changes within meaningful ranges, enhancing reliability and flexibility in AI explanations.

Contribution

CEMSP is a novel, general framework that models counterfactual explanation generation as a satisfiability problem, improving robustness and flexibility over existing methods.

Findings

CEMSP produces more stable explanations across similar instances.

It maintains flexibility while enhancing robustness.

Experimental results show improved explanation reliability.

Abstract

Counterfactual explanations (CFEs) exemplify how to minimally modify a feature vector to achieve a different prediction for an instance. CFEs can enhance informational fairness and trustworthiness, and provide suggestions for users who receive adverse predictions. However, recent research has shown that multiple CFEs can be offered for the same instance or instances with slight differences. Multiple CFEs provide flexible choices and cover diverse desiderata for user selection. However, individual fairness and model reliability will be damaged if unstable CFEs with different costs are returned. Existing methods fail to exploit flexibility and address the concerns of non-robustness simultaneously. To address these issues, we propose a conceptually simple yet effective solution named Counterfactual Explanations with Minimal Satisfiable Perturbations (CEMSP). Specifically, CEMSP constrains changing values of abnormal features with the help of their semantically meaningful normal ranges. For efficiency, we model the problem as a Boolean satisfiability problem to modify as few features as possible. Additionally, CEMSP is a general framework and can easily accommodate more practical requirements, e.g., casualty and actionability. Compared to existing methods, we conduct comprehensive experiments on both synthetic and real-world datasets to demonstrate that our method provides more robust explanations while preserving flexibility.