Sep 2026· IEEE Transactions on Knowledge and Data Engineering· Vol 38, pp. 6018-6037· 0 citations· 130 references
Abstract
Algorithmic fairness and explainability are foundational pillars of responsible AI. Although often studied independently, their interplay is increasingly recognized as crucial for diagnosing and mitigating bias in machine learning systems. We first introduce two systematic taxonomies: one for algorithmic fairness and one for explainable AI, to organize the landscape of existing work across diverse tasks (classification, ranking, and recommendation) and data modalities (tabular, graph). Next, we categorize the use of explanations in fairness efforts into three main functions: (a) detecting and understanding the causes of unfairness, (b) defining enhanced fairness metrics, and (c) designing mitigation strategies. In addition, we examine how explanation methods themselves can be biased, underscoring the need to evaluate fairness for explanations. Finally, we identify open research challenges and outline promising directions for future research at the intersection of fairness and explainability.
Machine learning models are increasingly deployed in high-stakes domains, raising concerns about both privacy and fairness. Differential Privacy (DP) has become a gold standard for privacy-preserving data analysis, while fairness-aware mechanisms aim to mitigate discrimination against underrepresented groups. However, these objectives can conflict: DP often amplifies disparities across demographic groups, and little is known about whether established fairness interventions remain effective under DP constraints. In this work, we present, to our knowledge, the first systematic evaluation of fairness interventions on differentially private synthetic tabular data. Our benchmark centers on the Adaptive Iterative Mechanism (AIM), identified as the state-of-the-art marginal-based DP synthesizer in recent KDD & VLDB 2025 tutorials by Cormode et al. We thus evaluate fairness interventions across four datasets, multiple group fairness metrics, and three categories of mitigation strategies (pre-processing, in-processing, and post-processing) under a wide range of privacy budgets. We compare four pipeline configurations: (Baseline) training on original data; (DP-only) training on DP synthetic data; (Fair-only) applying fairness mechanisms on original data; and (DP+Fair) combining fairness mechanisms with DP synthetic data. Our results demonstrate that while DP alone can degrade both utility and fairness, applying fairness interventions can partially restore equitable outcomes. Among them, post-processing methods tend to provide more stable fairness–utility trade-offs across privacy budgets and synthesizers, achieving strong fairness improvements while preserving competitive utility relative to other intervention stages. We release all code, data, and experimental artifacts in an open-source repository (https://github.com/vinicius-verona/dp-fair-intervention-benchmark) to ensure full reproducibility and to support future research on the privacy-fairness-utility trade-off.
Vin'icius Gabriel Angelozzi, H. H. Arcolezi· Proceedings on Privacy Enhan...· 0 citations
Algorithmic fairness methods are increasingly used to identify and mitigate bias in machine learning models, yet most approaches are evaluated in isolation and along single demographic axes. This limits practical guidance for selecting fairness strategies, where disparities may arise across intersectional subgroups and across multiple stages of the modeling lifecycle. This work presents FairSelect, a toolkit for systematically evaluating fairness mitigation strategies applied individually and in combination across preprocessing, inprocessing, and postprocessing stages. FairSelect supports multiple model architectures, intersectional subgroup evaluation, and comparison of fairness utility tradeoffs across baseline, single method, and multi level configurations. The framework was validated using synthetic clinical datasets designed to represent specific bias mechanisms and a real-world replication of two-year stroke risk prediction among patients with atrial fibrillation. Synthetic experiments showed that targeted fairness methods generally reduced intended subgroup disparities, while combined strategies produced larger average fairness improvements with modest utility tradeoffs. In the clinical prediction task, mitigation effects were highly variable, with some combinations improving both fairness and predictive performance while others were ineffective or counterproductive. These findings demonstrate that fairness interventions interact in nonadditive and context dependent ways. FairSelect provides a practical framework for systematically identifying fairness strategies that improve subgroup equity while preserving model performance in clinical machine learning.
Nick Souligne, Isabella Mixton-Garcia, V. Subbian· 0 citations
Normative datasets are often used to train and align AI systems, but the norms they contain can function as action-guiding patterns rather than neutral moral knowledge. We propose treating the AI system as a proxy actor and test whether dataset-level norms can shift it away from its baseline safety behavior when it faces high-conflict dilemmas. We make three contributions. First, we demonstrate in controlled experiments that norm-breaking fine-tuning yields norm-divergent actions justified by self-interested rationales, suggesting a systematic shift in patterns of justification. Second, we establish a practical audit trail linking downstream justifications to upstream norms using mixed methods. Third, we show that system prompts can both suppress and elicit these patterns. We conducted experiments on three models (LLaMA-3.2-11B, Qwen-3.5-9B, and Pixtral-12B) using Low-Rank Adaptation (LoRA) fine-tuning on Social Chemistry 101 Fairness/Cheating (norm-following vs. norm-breaking) with prompt steering. Across all three models, we find that norm-breaking fine-tuning shifts the model's default rationale style from safety compliance to instrumental self-interest, whereas system prompts can override this behavior. Our results support a distributed view of alignment in which observed behavior depends jointly on training data, fine-tuning, and prompting, motivating norm-aware documentation and rationale logging for contestable oversight.
L. H. Nguyen, Brice Valentin Kok-Shun, Guangyu Du et al.· 0 citations
Data-driven software systems are increasingly deployed in high-stakes socio-economic domains, from criminal justice to financial lending. However, these systems often exhibit individual discrimination---unjustified disparities in which a program yields different outcomes for similar individuals who differ only in their protected attributes (e.g., race, gender, age). While existing research has focused on detecting and quantifying these bugs, there remains a critical lack of principled mechanisms to explain and localize individual fairness bugs. Current explanation techniques are largely designed for single-input decisions rather than the relational nature of discrimination, which inherently involves a comparison between an original and a counterfactual pair. We present REMI, a framework for the automated localization, explanation, and mitigation of individual discrimination. Inspired by loop-invariant synthesis in formal methods, we treat counterfactual fairness as a relational invariant discovery problem. We introduce a bidirectional relational explanation framework that learns over paired examples $(x, x')$ to identify regions of the input space where fairness is violated. Unlike traditional one-way implication pairs used in invariant inference, our approach enforces bidirectional constraints: requiring identical outcomes for both original and counterfactual samples. REMI utilizes three data-alignment techniques to infer interpretable rule-based models that act as"fairness invariants."These rules serve as guardrails to selectively block or relabel unfair predictions without requiring model retraining. Our evaluation on symbolic and neural network programs demonstrates that REMI localizes ground-truth fairness bugs in over 83% of cases, significantly outperforming state-of-the-art baselines and reducing discriminatory decisions in black-box models by up to 70%.
Ranit Debnath Akash, Ashish Kumar, Gang Tan et al.· 0 citations
For a fair society, decisions impacting people's lives must be taken fairly, irrespective of their protected characteristics. To achieve algorithmic fairness, appropriate measures must guide machine learning methods toward fair(er) decision recommendations. There are numerous fairness metrics for classification; however, for regression the literature is still developing. Most fair machine learning methods combine fairness and error in a single objective. Instead, we apply multi-objective optimization, allowing the fairness metrics to be optimized alongside the error. We develop a transition from classification to regression for symbolic regression by first discretizing the task into a fixed number of classes, progressively increasing their number, and finally by using full regression fairness metrics. We demonstrate the framework on the Law School Admission Council dataset. While fairness objectives do not appear to be significantly different among methods, accuracy is clearly better when applying full regression fairness.
Michel El Saliby, David Ibehej, Nicole Vergara et al.· Proceedings of the Genetic a...· 0 citations
The growing adoption of artificial intelligence in the financial sector has intensified concerns regarding unfair discrimination across diverse systems. Under increasing regulatory and accountability pressures, ensuring fairness and transparency in AI-driven decision-making has become a critical challenge. Our aim is to map how the literature addresses fairness in finance, the metrics employed, the financial contexts considered, and the effectiveness of mitigation techniques. This review also seeks to synthesize existing knowledge, identify methodological gaps, and provide guidance for future research and policy development. We considered peer-reviewed articles focused on AI in finance and fairness, prioritizing studies from 2023–2026 or from 2016–2022 with at least 100 citations. The papers were collected from CAPES, Elsevier, Google Scholar, Scopus, Web of Science, and gray literature such as ArXiv. We performed automated screening, AI-based refinement (Gemini, Perplexity, Copilot), and systematic extraction of technical dimensions using a results spreadsheet for thematic categorization of articles considered relevant. We identified 17,019 records, resulting in a final corpus of 99 articles for the synthesis of the work and 45 supporting documents for contextualizing finance and equity. We answered seven research questions related to datasets, machine learning, and equity approaches, as well as evaluation metrics. We found a predominance of studies on credit risk and credit scoring. There was a consolidation of technical approaches and identification of the need for standardized metrics, with greater emphasis on intersectionality and causality for sensitive data.
Marcelo Wecchi, Lilian Berton· AI and Ethics· 0 citations