Adaptive Adversarial Training Does Not Increase Recourse Costs

TL;DR

This paper investigates whether adaptive adversarial training increases the cost of algorithmic recourse and finds that it does not, suggesting a way to achieve robustness without additional recourse costs.

Contribution

It demonstrates that adaptive adversarial training enhances model robustness without significantly raising recourse costs, unlike traditional adversarial training.

Findings

Adaptive adversarial training does not substantially increase recourse costs.

Model robustness improves with adaptive adversarial training.

Recourse costs remain manageable under adaptive adversarial training.

Abstract

Recent work has connected adversarial attack methods and algorithmic recourse methods: both seek minimal changes to an input instance which alter a model's classification decision. It has been shown that traditional adversarial training, which seeks to minimize a classifier's susceptibility to malicious perturbations, increases the cost of generated recourse; with larger adversarial training radii correlating with higher recourse costs. From the perspective of algorithmic recourse, however, the appropriate adversarial training radius has always been unknown. Another recent line of work has motivated adversarial training with adaptive training radii to address the issue of instance-wise variable adversarial vulnerability, showing success in domains with unknown attack radii. This work studies the effects of adaptive adversarial training on algorithmic recourse costs. We establish that the improvements in model robustness induced by adaptive adversarial training show little effect on algorithmic recourse costs, providing a potential avenue for affordable robustness in domains where recoursability is critical.