面向无需人口统计先验的无害罗尔斯公平性
Towards Harmless Rawlsian Fairness Regardless of Demographic Prior
AI总结：
针对无人口统计先验的公平性需求，提出VFair方法，通过最小化训练损失方差实现无害罗尔斯公平，实验表明回归任务公平性显著提升，分类任务效果有限。
摘要：
由于隐私和安全方面的担忧，群体公平领域的最新进展提倡在训练模型时不依赖人口统计信息。然而，大多数方法仍然需要人口统计的先验知识。在本研究中，我们探讨了在训练集未提供任何人口统计先验的情况下，实现公平性且不损害其效用的可能性，即“无害罗尔斯公平性”。我们确认，这种无人口统计先验的公平性要求本质上促使训练损失呈现狄拉克δ分布。为此，我们提出了一种简单但有效的方法，名为VFair，用于在经验损失的最优集合内最小化训练损失的方差。该问题通过一种定制的动态更新方法进行优化，该方法在损失和梯度两个维度上运作，引导模型走向相对更公平的解，同时保持其完整的效用。我们的实验结果表明，回归任务（文献中相对未被充分探索）通过VFair可以实现显著的公平性提升，而无需任何先验；而分类任务通常则不然，这是因为其量化的效用度量。我们方法的实现公开于https://github.com/wxqpxw/VFair。
英文摘要：
Due to privacy and security concerns, recent advancements in group fairness advocate for model training regardless of demographic information. However, most methods still require prior knowledge of demographics. In this study, we explore the potential for achieving fairness without compromising its utility when no prior demographics are provided to the training set, namely \emph{harmless Rawlsian fairness}. We ascertain that such a fairness requirement with no prior demographic information essential promotes training losses to exhibit a Dirac delta distribution. To this end, we propose a simple but effective method named VFair to minimize the variance of training losses inside the optimal set of empirical losses. This problem is then optimized by a tailored dynamic update approach that operates in both loss and gradient dimensions, directing the model towards relatively fairer solutions while preserving its intact utility. Our experimental findings indicate that regression tasks, which are relatively unexplored from literature, can achieve significant fairness improvement through VFair regardless of any prior, whereas classification tasks usually do not because of their quantized utility measurements. The implementation of our method is publicly available at \url{https://github.com/wxqpxw/VFair}.
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。
