Relevance-aware Algorithmic Recourse

TL;DR

This paper introduces Relevance-Aware Algorithmic Recourse (RAAR), a framework that incorporates relevance into generating actionable model explanations, improving efficiency and reducing costs in regression tasks.

Contribution

The paper proposes a novel relevance-aware framework for algorithmic recourse in regression, addressing the limitation of treating all domain values equally.

Findings

Relevance influences the quality of recourses.

RAAR achieves comparable results to baselines.

Relevance-aware recourses are more efficient and cost-effective.

Abstract

As machine learning continues to gain prominence, transparency and explainability are increasingly critical. Without an understanding of these models, they can replicate and worsen human bias, adversely affecting marginalized communities. Algorithmic recourse emerges as a tool for clarifying decisions made by predictive models, providing actionable insights to alter outcomes. They answer, 'What do I have to change?' to achieve the desired result. Despite their importance, current algorithmic recourse methods treat all domain values equally, which is unrealistic in real-world settings. In this paper, we propose a novel framework, Relevance-Aware Algorithmic Recourse (RAAR), that leverages the concept of relevance in applying algorithmic recourse to regression tasks. We conducted multiple experiments on 15 datasets to outline how relevance influences recourses. Results show that relevance contributes algorithmic recourses comparable to well-known baselines, with greater efficiency and lower relative costs.