归一化以实现公平：回归机器学习问题中一种简单的公平性归一化技术
Normalise for Fairness: A Simple Normalisation Technique for Fairness in Regression Machine Learning Problems
AI总结：
本文提出一种基于归一化的简单公平性方法FaiReg及其混合变体FaiRegH，用于回归任务，有效减少标签偏差导致的不公平，且不显著损害预测性能，并在First Impressions数据集上验证了其优越性。
摘要：
算法和机器学习（ML）日益影响日常生活和多个决策过程，其中ML因可扩展性或优越性能而具有优势。在此类应用中，公平性至关重要，模型不应基于种族、性别或其他受保护群体歧视其结果。这对于影响非常敏感话题的模型尤为关键，如面试邀请或累犯预测。与二分类问题相比，公平性在回归问题中研究较少；因此，我们提出了一种基于归一化的简单而有效的方法（FaiReg），它最小化了回归问题中不公平性的影响，尤其是标签偏差带来的不公平性。我们对该方法进行了理论分析，并与两种标准的公平性方法（即数据平衡和对抗训练）进行了实证比较。我们还提出了一种混合形式（FaiRegH），将所提出的方法与数据平衡相结合，旨在同时应对标签偏差和采样偏差。实验在多模态数据集First Impressions（FI）上进行，使用多种标签，即大五人格预测和面试筛选评分。结果表明，在减少不公平性影响方面，该方法优于数据平衡，且不会像对抗训练那样严重损害原始问题的性能。公平性基于Equal Accuracy（EA）和Statistical Parity（SP）约束进行评估。实验设置同时增强了多个受保护变量的公平性。
英文摘要：
Algorithms and Machine Learning (ML) are increasingly affecting everyday life and several decision-making processes, where ML has an advantage due to scalability or superior performance. Fairness in such applications is crucial, where models should not discriminate their results based on race, gender, or other protected groups. This is especially crucial for models affecting very sensitive topics, like interview invitation or recidivism prediction. Fairness is not commonly studied for regression problems compared to binary classification problems; hence, we present a simple, yet effective method based on normalisation (FaiReg), which minimises the impact of unfairness in regression problems, especially due to labelling bias. We present a theoretical analysis of the method, in addition to an empirical comparison against two standard methods for fairness, namely data balancing and adversarial training. We also include a hybrid formulation (FaiRegH), merging the presented method with data balancing, in an attempt to face labelling and sampling biases simultaneously. The experiments are conducted on the multimodal dataset First Impressions (FI) with various labels, namely Big-Five personality prediction and interview screening score. The results show the superior performance of diminishing the effects of unfairness better than data balancing, also without deteriorating the performance of the original problem as much as adversarial training. Fairness is evaluated based on the Equal Accuracy (EA) and Statistical Parity (SP) constraints. The experiments present a setup that enhances the fairness for several protected variables simultaneously.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
