The concern about hidden discrimination in machine learning models is growing, as their widespread real-world applications increasingly impact human lives. Various techniques, including commonly used group fairness measures and several fairness-aware ensemble-based methods, have been developed to enhance fairness. However, existing fairness measures typically focus on only one aspect -- either group or individual fairness, and the compatibility difficulty among these measures indicates a possibility of remaining biases even when one of them is satisfied. Moreover, existing mechanisms to boost fairness usually present empirical results to show validity, yet few of them discuss whether fairness can be boosted with certain theoretical guarantees. To address these issues, we propose a fairness quality measure named `discriminative risk' by only perturbing protected attributes in instances, to express both individual and group fairness aspects. Furthermore, we investigate its properties and establish the first- and second-order oracle bounds and their relaxations, which show that fairness is possibly improved via ensemble combination with margin-dependent bounds. The analysis is suitable for both binary and multi-class classification. A few ensemble pruning methods are also proposed to utilise our proposed measure and obtain both accurate and fair sub-ensembles; comprehensive experiments are conducted to evaluate the effectiveness of the proposed fairness measure and pruning methods.

翻译：随着机器学习模型在现实世界中的广泛应用日益影响人类生活，对其潜在歧视问题的关注也日益增长。为增强公平性，已发展出多种技术，包括常用的群体公平性度量以及若干基于公平意识的集成方法。然而，现有的公平性度量通常仅关注单一层面——群体公平性或个体公平性，且这些度量之间的兼容性困难表明，即使满足其中一种度量，仍可能存在残余偏差。此外，现有提升公平性的机制多通过实证结果展示有效性，但鲜有研究探讨公平性是否能在特定理论保证下得到提升。为解决这些问题，我们提出一种名为“判别风险”的公平性质量度量，该方法仅通过扰动样本中的受保护属性来同时表达个体与群体公平性层面。进一步，我们研究了该度量的性质，并建立了一阶与二阶Oracle界及其松弛形式，这些结果表明通过集成组合并借助间隔依赖的界，公平性有可能得到提升。该分析适用于二分类与多分类任务。我们还提出了若干集成剪枝方法，以利用所提出的度量来获得既准确又公平的子集成；通过全面的实验评估了所提出的公平性度量与剪枝方法的有效性。