On the Adversarial Robustness of Causal Algorithmic Recourse

TL;DR

This paper investigates the robustness of algorithmic recourse against adversarial perturbations, proposing methods to generate and improve robust recourse for linear and differentiable classifiers.

Contribution

It formulates the adversarially robust recourse problem and introduces methods to generate and enhance robust recourse, especially through classifier regularization.

Findings

Minimally costly recourse methods lack robustness.

Proposed methods generate adversarially robust recourse.

Regularizing classifiers improves robustness of recourse.

Abstract

Algorithmic recourse seeks to provide actionable recommendations for individuals to overcome unfavorable classification outcomes from automated decision-making systems. Recourse recommendations should ideally be robust to reasonably small uncertainty in the features of the individual seeking recourse. In this work, we formulate the adversarially robust recourse problem and show that recourse methods that offer minimally costly recourse fail to be robust. We then present methods for generating adversarially robust recourse for linear and for differentiable classifiers. Finally, we show that regularizing the decision-making classifier to behave locally linearly and to rely more strongly on actionable features facilitates the existence of adversarially robust recourse.