Explaining Black-Box Models through Counterfactuals

TL;DR

This paper introduces CounterfactualExplanations.jl, a Julia package for generating counterfactual explanations and algorithmic recourse to interpret black-box models, emphasizing usability, customization, and extensibility.

Contribution

The paper presents a new Julia package that provides customizable and extensible tools for generating counterfactual explanations and recourse for black-box models.

Findings

The package effectively generates realistic counterfactual explanations.

It is user-friendly and easily customizable.

The package supports diverse counterfactual generation methods.

Abstract

We present CounterfactualExplanations.jl: a package for generating Counterfactual Explanations (CE) and Algorithmic Recourse (AR) for black-box models in Julia. CE explain how inputs into a model need to change to yield specific model predictions. Explanations that involve realistic and actionable changes can be used to provide AR: a set of proposed actions for individuals to change an undesirable outcome for the better. In this article, we discuss the usefulness of CE for Explainable Artificial Intelligence and demonstrate the functionality of our package. The package is straightforward to use and designed with a focus on customization and extensibility. We envision it to one day be the go-to place for explaining arbitrary predictive models in Julia through a diverse suite of counterfactual generators.