一篇新论文提出了一种新颖的分类器公平性定义，该定义考虑了特征之间的约束。作者认为，如果一个决策具有“公平的解释”，则该决策是公平的，公平的解释被定义为排除受保护属性的、考虑特征约束的决策的素蕴涵原因。该研究探讨了不同公平性定义之间的关系，并分析了测试分类器公平性的计算复杂性，强调了忽略约束如何显著改变公平性评估。 AI
影响 引入了评估分类器公平性的新理论框架，可能影响对AI系统偏见的审计方式。
排序理由 关于分类器公平性的学术论文，包含新颖的定义和复杂性分析。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2605.00592v1 Announce Type: new Abstract: In Machine Learning, an accepted definition of fairness of a decision taken by a classifier is that it should not depend on protected features, such as gender. Unfortunately, when constraints exist between features, such dependencie…
In Machine Learning, an accepted definition of fairness of a decision taken by a classifier is that it should not depend on protected features, such as gender. Unfortunately, when constraints exist between features, such dependencies can be obscured by the constraints. To avoid t…