arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.01385cs.LGcs.AI
收藏

无需访问敏感属性的公平分类超参数调优

Hyper-parameter Tuning for Fair Classification without Sensitive Attribute Access

New York University(纽约大学);University of Maryland College Park(马里兰大学帕克分校);JP Morgan Chase AI Research(摩根大通人工智能研究院)

Akshaj Kumar Veldanda, Ivan Brugere, Sanghamitra Dutta, Alan Mishler, Siddharth Garg

AI总结:

提出Antigone框架,通过训练有偏分类器生成伪敏感属性,在训练和验证数据均无敏感属性的情况下实现公平分类器的超参数调优,在理论和实验上证明该方法能在平均准确率约束下最大化公平性。

摘要:

公平机器学习方法旨在训练模型,以平衡在由种族和性别等敏感属性定义的人口学子群体上的模型性能。尽管通常假设在训练期间敏感属性是已知的,但由于隐私和其他实际操作问题,在实践中它们可能不可用。最近的工作致力于在没有训练数据敏感属性的情况下训练公平模型。然而,这些方法需要大量的超参数调优才能取得良好结果,因此假设在验证数据上敏感属性是已知的。然而,这种假设可能同样不切实际。在此,我们提出了 Antigone,一个在训练或验证数据均无法访问敏感属性的情况下训练公平分类器的框架。相反,我们通过训练一个有偏分类器,并将该分类器错误(正确)标记的样本作为少数(多数)群体的代理,从而在验证数据上生成伪敏感属性。由于即使存在噪声敏感属性信息,人口学平权、机会均等和子群体准确率等公平性指标也能在比例常数范围内被估计,我们在理论和实验上证明了这些代理标签可用于在平均准确率约束下最大化公平性。我们结果的关键在于一种原则性方法,该方法以完全无监督的方式(即无需访问真实标签敏感属性)选择有偏分类器的超参数,从而最小化使用噪声与真实敏感标签估计的公平性之间的差距。

英文摘要:

Fair machine learning methods seek to train models that balance model performance across demographic subgroups defined over sensitive attributes like race and gender. Although sensitive attributes are typically assumed to be known during training, they may not be available in practice due to privacy and other logistical concerns. Recent work has sought to train fair models without sensitive attributes on training data. However, these methods need extensive hyper-parameter tuning to achieve good results, and hence assume that sensitive attributes are known on validation data. However, this assumption too might not be practical. Here, we propose Antigone, a framework to train fair classifiers without access to sensitive attributes on either training or validation data. Instead, we generate pseudo sensitive attributes on the validation data by training a biased classifier and using the classifier's incorrectly (correctly) labeled examples as proxies for minority (majority) groups. Since fairness metrics like demographic parity, equal opportunity and subgroup accuracy can be estimated to within a proportionality constant even with noisy sensitive attribute information, we show theoretically and empirically that these proxy labels can be used to maximize fairness under average accuracy constraints. Key to our results is a principled approach to select the hyper-parameters of the biased classifier in a completely unsupervised fashion (meaning without access to ground truth sensitive attributes) that minimizes the gap between fairness estimated using noisy versus ground-truth sensitive labels.

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。

↑