The Illusion of Fairness: Auditing Fairness Interventions with Audit Studies

• URL: http://arxiv.org/abs/2507.02152v1
• Date: Wed, 02 Jul 2025 21:15:56 GMT
• Title: The Illusion of Fairness: Auditing Fairness Interventions with Audit Studies
• Authors: Disa Sariola, Patrick Button, Aron Culotta, Nicholas Mattei,
• Abstract summary: We investigate how data from audit studies can be used to improve our ability to both train and evaluate automated hiring algorithms.<n>We find that such data reveals cases where the common fairness intervention method of equalizing base rates across classes appears to achieve parity using traditional measures.
• Score: 12.7910460546047
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Artificial intelligence systems, especially those using machine learning, are being deployed in domains from hiring to loan issuance in order to automate these complex decisions. Judging both the effectiveness and fairness of these AI systems, and their human decision making counterpart, is a complex and important topic studied across both computational and social sciences. Within machine learning, a common way to address bias in downstream classifiers is to resample the training data to offset disparities. For example, if hiring rates vary by some protected class, then one may equalize the rate within the training set to alleviate bias in the resulting classifier. While simple and seemingly effective, these methods have typically only been evaluated using data obtained through convenience samples, introducing selection bias and label bias into metrics. Within the social sciences, psychology, public health, and medicine, audit studies, in which fictitious ``testers'' (e.g., resumes, emails, patient actors) are sent to subjects (e.g., job openings, businesses, doctors) in randomized control trials, provide high quality data that support rigorous estimates of discrimination. In this paper, we investigate how data from audit studies can be used to improve our ability to both train and evaluate automated hiring algorithms. We find that such data reveals cases where the common fairness intervention method of equalizing base rates across classes appears to achieve parity using traditional measures, but in fact has roughly 10% disparity when measured appropriately. We additionally introduce interventions based on individual treatment effect estimation methods that further reduce algorithmic discrimination using this data.

Related papers

• Simulating Biases for Interpretable Fairness in Offline and Online Classifiers [0.35998666903987897]
  Mitigation methods are critical to ensure that model outcomes are adjusted to be fair.<n>We develop a framework for synthetic dataset generation with controllable bias injection.<n>In experiments, both offline and online learning approaches are employed.
  arXiv  Detail & Related papers  (2025-07-14T11:04:24Z)
• Underrepresentation, Label Bias, and Proxies: Towards Data Bias Profiles for the EU AI Act and Beyond [42.710392315326104]
  We present three common data biases and study their individual and joint effect on algorithmic discrimination.<n>We develop dedicated mechanisms to detect specific types of bias, and combine them into a preliminary construct we refer to as the Data Bias Profile (DBP)<n>This initial formulation serves as a proof of concept for how different bias signals can be systematically documented.
  arXiv  Detail & Related papers  (2025-07-09T15:52:11Z)
• On the Interconnections of Calibration, Quantification, and Classifier Accuracy Prediction under Dataset Shift [58.91436551466064]
  This paper investigates the interconnections among three fundamental problems, calibration, and quantification, under dataset shift conditions.<n>We show that access to an oracle for any one of these tasks enables the resolution of the other two.<n>We propose new methods for each problem based on direct adaptations of well-established methods borrowed from the other disciplines.
  arXiv  Detail & Related papers  (2025-05-16T15:42:55Z)
• Outlier Detection Bias Busted: Understanding Sources of Algorithmic Bias through Data-centric Factors [28.869581543676947]
  unsupervised outlier detection (OD) has numerous applications in finance, security, etc. This work aims to shed light on the possible sources of unfairness in OD by auditing detection models under different data-centric factors. We find that the OD algorithms under the study all exhibit fairness pitfalls, although differing in which types of data bias they are more susceptible to.
  arXiv  Detail & Related papers  (2024-08-24T20:35:32Z)
• D-BIAS: A Causality-Based Human-in-the-Loop System for Tackling Algorithmic Bias [57.87117733071416]
  We propose D-BIAS, a visual interactive tool that embodies human-in-the-loop AI approach for auditing and mitigating social biases. A user can detect the presence of bias against a group by identifying unfair causal relationships in the causal network. For each interaction, say weakening/deleting a biased causal edge, the system uses a novel method to simulate a new (debiased) dataset.
  arXiv  Detail & Related papers  (2022-08-10T03:41:48Z)
• Understanding Unfairness in Fraud Detection through Model and Data Bias Interactions [4.159343412286401]
  We argue that algorithmic unfairness stems from interactions between models and biases in the data. We study a set of hypotheses regarding the fairness-accuracy trade-offs that fairness-blind ML algorithms exhibit under different data bias settings.
  arXiv  Detail & Related papers  (2022-07-13T15:18:30Z)
• Statistical discrimination in learning agents [64.78141757063142]
  Statistical discrimination emerges in agent policies as a function of both the bias in the training population and of agent architecture. We show that less discrimination emerges with agents that use recurrent neural networks, and when their training environment has less bias.
  arXiv  Detail & Related papers  (2021-10-21T18:28:57Z)
• Automatic Fairness Testing of Neural Classifiers through Adversarial Sampling [8.2868128804393]
  We propose a scalable and effective approach for systematically searching for discriminative samples. Compared with state-of-the-art methods, our approach only employs lightweight procedures like gradient computation and clustering. The retrained models reduce discrimination by 57.2% and 60.2% respectively on average.
  arXiv  Detail & Related papers  (2021-07-17T03:47:08Z)
• Can Active Learning Preemptively Mitigate Fairness Issues? [66.84854430781097]
  dataset bias is one of the prevailing causes of unfairness in machine learning. We study whether models trained with uncertainty-based ALs are fairer in their decisions with respect to a protected class. We also explore the interaction of algorithmic fairness methods such as gradient reversal (GRAD) and BALD.
  arXiv  Detail & Related papers  (2021-04-14T14:20:22Z)
• Domain-Incremental Continual Learning for Mitigating Bias in Facial Expression and Action Unit Recognition [5.478764356647437]
  We propose the novel usage of Continual Learning (CL) as a potent bias mitigation method to enhance the fairness of FER systems. We compare different non-CL-based and CL-based methods for their classification accuracy and fairness scores on expression recognition and Action Unit (AU) detection tasks. Our experimental results show that CL-based methods, on average, outperform other popular bias mitigation techniques on both accuracy and fairness metrics.
  arXiv  Detail & Related papers  (2021-03-15T18:22:17Z)
• Coping with Mistreatment in Fair Algorithms [1.2183405753834557]
  We study the algorithmic fairness in a supervised learning setting and examine the effect of optimizing a classifier for the Equal Opportunity metric. We propose a conceptually simple method to mitigate this bias. We rigorously analyze the proposed method and evaluate it on several real world datasets demonstrating its efficacy.
  arXiv  Detail & Related papers  (2021-02-22T03:26:06Z)
• Towards Model-Agnostic Post-Hoc Adjustment for Balancing Ranking Fairness and Algorithm Utility [54.179859639868646]
  Bipartite ranking aims to learn a scoring function that ranks positive individuals higher than negative ones from labeled data. There have been rising concerns on whether the learned scoring function can cause systematic disparity across different protected groups. We propose a model post-processing framework for balancing them in the bipartite ranking scenario.
  arXiv  Detail & Related papers  (2020-06-15T10:08:39Z)
• Bias in Multimodal AI: Testbed for Fair Automatic Recruitment [73.85525896663371]
  We study how current multimodal algorithms based on heterogeneous sources of information are affected by sensitive elements and inner biases in the data. We train automatic recruitment algorithms using a set of multimodal synthetic profiles consciously scored with gender and racial biases. Our methodology and results show how to generate fairer AI-based tools in general, and in particular fairer automated recruitment systems.
  arXiv  Detail & Related papers  (2020-04-15T15:58:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.