不重训练也不改模型，用反事实平均修复预测偏见
CAFP: A Post-Processing Framework for Group Fairness via Counterfactual Model Averaging
- 对每个输入生成敏感属性翻转的反事实样本，平均真实与反事实预测结果
- 理论证明可消除敏感属性直接影响，降低预测与属性的互信息
- 适合无法修改模型或访问敏感属性的现实系统，如信贷评分
在信贷评分、医疗和刑事司法等敏感领域，确保机器学习预测的公平性至关重要。尽管许多公平性方法依赖数据预处理或训练时的算法约束，但这些方法通常需要完全控制模型架构并访问受保护属性信息，在实际系统中往往不可行。本文提出一种无需重训练或修改原分类器的模型无关后处理方法——反事实平均公平预测（CAFP）。CAFP通过生成每个输入在敏感属性翻转后的反事实版本，将模型在真实与反事实实例上的预测结果进行平均。我们提供了理论分析，表明该方法可消除对受保护属性的直接依赖，降低预测与敏感属性间的互信息，并在温和假设下，证明其引入的偏差扭曲有界。进一步地，我们证明了在满足条件下，CAFP可实现完全人口均等，并将等化机会差距减少至少一半的平均反事实偏见。
原文摘要 · Abstract (English)
Ensuring fairness in machine learning predictions is a critical challenge, especially when models are deployed in sensitive domains such as credit scoring, healthcare, and criminal justice. While many fairness interventions rely on data preprocessing or algorithmic constraints during training, these approaches often require full control over the model architecture and access to protected attribute information, which may not be feasible in real-world systems. In this paper, we propose Counterfactual Averaging for Fair Predictions (CAFP), a model-agnostic post-processing method that mitigates unfair influence from protected attributes without retraining or modifying the original classifier. CAFP operates by generating counterfactual versions of each input in which the sensitive attribute is flipped, and then averaging the model's predictions across factual and counterfactual instances. We provide a theoretical analysis of CAFP, showing that it eliminates direct dependence on the protected attribute, reduces mutual information between predictions and sensitive attributes, and provably bounds the distortion introduced relative to the original model. Under mild assumptions, we further show that CAFP achieves perfect demographic parity and reduces the equalized odds gap by at least half the average counterfactual bias.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。