一句话：从博弈论交互视角解释敏感信息编码，提出 HIFI 缓解算法偏见的 in-processing 方法。
机器学习算法在关键社会领域决策中常继承并放大训练数据中的偏见，导致不公平结果。现有工作多从统计角度分析偏见根源，却未说明引发歧视决策的敏感信息如何被模型编码。本文从博弈论视角解释并缓解算法偏见：利用 Harsanyi 交互对多种公平指标隐含的敏感信息进行数学刻画，并据此提出 in-processing 偏见缓解方法 HIFI。评估涵盖 11 种先进方法、5 个真实数据集、4 种公平准则和 5 个 ML 性能指标，并考察多个受保护属性的交叉公平。结果表明，HIFI 在公平性提升与公平-性能权衡上优于现有 in-processing 方法，同时显著减少个体公平违规。
Machine Learning (ML) algorithms are increasingly used in decision-making process across various social-critical domains, but they often somewhat inherit and amplify bias from their training data, leading to unfair and unethical outcomes. This issue highlights the urgent need for effective methods to detect, explain, and mitigate bias to ensure the fairness of ML systems. Previous studies are prone to analyze the root causes of algorithmic bias from a statistical perspective. However, to the best of our knowledge, none of them has discussed how sensitive information inducing the final discriminatory decision is encoded by ML models. In this work, we attempt to explain and mitigate algorithmic bias from a game-theoretic view. We mathematically decode an essential and common component of sensitive information implicitly defined by various fairness metrics with Harsanyi interactions, and on this basis, we propose an in-processing method HIFI for bias mitigation. We conduct an extensive evaluation of HIFI with 11 state-of-the-art methods, 5 real-world datasets, 4 fairness criteria, and 5 ML performance metrics, while also considering intersectional fairness for multiple protected attributes. The results show that HIFI surpasses state-of-the-art in-processing methods in terms of fairness improvement and fairness-performance trade-off, and also achieves notable effectiveness in reducing violations of individual fairness simultaneously.
DOI 原文 ·
@inproceedings{paperbot8682,
title = {HIFI: Explaining and Mitigating Algorithmic Bias Through the Lens of Game-Theoretic Interactions},
author = {Lingfeng Zhang and Zhaohui Wang and Yueling Zhang and Min Zhang and Jiangtao Wang},
booktitle = {ICSE},
year = {2025},
doi = {10.1109/icse55347.2025.00221}
}