A Causal Perspective on Meaningful and Robust Algorithmic Recourse
This addresses the issue of unreliable recourse for stakeholders in ML systems, offering a causal perspective to enhance robustness and meaningfulness, though it builds incrementally on prior work.
The paper tackles the problem of algorithmic recourse explanations that may not improve the underlying target, proposing meaningful algorithmic recourse (MAR) to ensure actions improve both prediction and target, and effective algorithmic recourse (EAR) for interventions on causes under assumptions.
Algorithmic recourse explanations inform stakeholders on how to act to revert unfavorable predictions. However, in general ML models do not predict well in interventional distributions. Thus, an action that changes the prediction in the desired way may not lead to an improvement of the underlying target. Such recourse is neither meaningful nor robust to model refits. Extending the work of Karimi et al. (2021), we propose meaningful algorithmic recourse (MAR) that only recommends actions that improve both prediction and target. We justify this selection constraint by highlighting the differences between model audit and meaningful, actionable recourse explanations. Additionally, we introduce a relaxation of MAR called effective algorithmic recourse (EAR), which, under certain assumptions, yields meaningful recourse by only allowing interventions on causes of the target.