Counterfactual Explanations and Algorithmic Recourses for Machine Learning: A Review

TL;DR

This review paper categorizes and evaluates various counterfactual explanation methods in machine learning, emphasizing their importance for trustworthy AI in high-impact sectors like finance and healthcare.

Contribution

It introduces a comprehensive rubric for assessing counterfactual explanation algorithms and provides a systematic comparison of existing approaches.

Findings

Most algorithms aim for interpretability and fidelity.

Counterfactual explanations align with legal standards in many countries.

Research gaps include scalability and robustness of explanations.

Abstract

Machine learning plays a role in many deployed decision systems, often in ways that are difficult or impossible to understand by human stakeholders. Explaining, in a human-understandable way, the relationship between the input and output of machine learning models is essential to the development of trustworthy machine learning based systems. A burgeoning body of research seeks to define the goals and methods of explainability in machine learning. In this paper, we seek to review and categorize research on counterfactual explanations, a specific class of explanation that provides a link between what could have happened had input to a model been changed in a particular way. Modern approaches to counterfactual explainability in machine learning draw connections to the established legal doctrine in many countries, making them appealing to fielded systems in high-impact areas such as finance and healthcare. Thus, we design a rubric with desirable properties of counterfactual explanation algorithms and comprehensively evaluate all currently proposed algorithms against that rubric. Our rubric provides easy comparison and comprehension of the advantages and disadvantages of different approaches and serves as an introduction to major research themes in this field. We also identify gaps and discuss promising research directions in the space of counterfactual explainability.