Causally Perturbed Fairness Testing

TL;DR

CausalFT is a framework that uses causal inference to improve fairness testing in AI models by guiding perturbations based on causal relationships, significantly enhancing bug detection and bias resilience.

Contribution

It introduces a causal inference-based approach to guide perturbations in fairness testing, outperforming correlation-based methods and serving as a versatile framework adaptable to various generators.

Findings

CausalFT improves fairness bug detection in over 93% of cases.

It outperforms correlation-based feature ranking in 64% of cases.

CausalFT enhances bias resilience across nearly all tested scenarios.

Abstract

To mitigate unfair and unethical discrimination over sensitive features (e.g., gender, age, or race), fairness testing plays an integral role in engineering systems that leverage AI models to handle tabular data. A key challenge therein is how to effectively reveal fairness bugs under an intractable sample size using perturbation. Much current work has been focusing on designing the test sample generators, ignoring the valuable knowledge about data characteristics that can help guide the perturbation and hence limiting their full potential. In this paper, we seek to bridge such a gap by proposing a generic framework of causally perturbed fairness testing, dubbed CausalFT. Through causal inference, the key idea of CausalFT is to extract the most directly and causally relevant non-sensitive feature to its sensitive counterpart, which can jointly influence the prediction of the label. Such a causal relationship is then seamlessly injected into the perturbation to guide a test sample generator. Unlike existing generator-level work, CausalFT serves as a higher-level framework that can be paired with diverse base generators. Extensive experiments on 1296 cases confirm that CausalFT can considerably improve arbitrary base generators in revealing fairness bugs over 93% of the cases with acceptable extra runtime overhead. Compared with a state-of-the-art approach that ranks the non-sensitive features solely based on correlation, CausalFT performs significantly better on 64% cases while being much more efficient. Further, CausalFT can better improve bias resilience in nearly all cases.