arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2207.04581cs.LGcs.CY
收藏

你的公平模型有多鲁棒?探索多样化公平策略的鲁棒性

How Robust is your Fair Model? Exploring the Robustness of Diverse Fairness Strategies

RMIT University(皇家墨尔本理工大学);University of Michigan(密歇根大学);ETH Zurich(苏黎世联邦理工学院);UNSW Sydney(新南威尔士大学悉尼分校);ARC Centre of Excellence for Automated Decision-Making and Society(自动化决策与社会卓越研究中心);School of Computing Technologies, RMIT University(皇家墨尔本理工大学计算技术学院);Department of Mathematics, University of Michigan(密歇根大学数学系);Department of Computer Science, ETH Zurich(苏黎世联邦理工学院计算机科学系);School of Computer Science and Engineering, UNSW Sydney(新南威尔士大学悉尼分校计算机科学与工程学院)

Edward Small, Wei Shao, Zeliang Zhang, Peihan Liu, Jeffrey Chan, Kacper Sokol, Flora Salim

AI总结:

本文提出了一种名为鲁棒性比率的新标准,通过在五个基准数据集上评估三种公平策略和四种公平定义,首次定量分析了公平优化策略对噪声的鲁棒性,为不同数据集选择合适的公平策略提供了指导。

摘要:

随着机器学习被引入高风险决策中,确保算法公平性已成为一个日益重要且亟待解决的问题。为此,人们提出了许多关于公平性的数学定义,并开发了多种优化技术,所有这些都是为了最大化某种特定的公平性概念。然而,公平的解决方案依赖于训练数据的质量,并且可能对噪声高度敏感。最近的研究表明,鲁棒性(模型在未见数据上表现良好的能力)在处理新问题时应采取何种策略方面起着重要作用,因此,衡量这些策略的鲁棒性已成为一个基本问题。在这项工作中,我们因此提出了一种新的标准来衡量各种公平优化策略的鲁棒性——鲁棒性比率。我们在五个基准公平数据集上进行了多次广泛的实验,使用了三种最流行的公平策略,并涉及四种最流行的公平性定义。我们的实验从经验上表明,尽管依赖阈值优化的公平方法在大多数情况下优于其他方法,但它们在所有评估的数据集中对噪声都非常敏感。这与另外两种方法形成对比,后两种方法在低噪声场景下公平性较差,但在高噪声场景下公平性更好。据我们所知,我们是首个定量评估公平优化策略鲁棒性的研究。这有可能为各种数据集选择最合适的公平策略提供指导方针。

英文摘要:

With the introduction of machine learning in high-stakes decision making, ensuring algorithmic fairness has become an increasingly important problem to solve. In response to this, many mathematical definitions of fairness have been proposed, and a variety of optimisation techniques have been developed, all designed to maximise a defined notion of fairness. However, fair solutions are reliant on the quality of the training data, and can be highly sensitive to noise. Recent studies have shown that robustness (the ability for a model to perform well on unseen data) plays a significant role in the type of strategy that should be used when approaching a new problem and, hence, measuring the robustness of these strategies has become a fundamental problem. In this work, we therefore propose a new criterion to measure the robustness of various fairness optimisation strategies - the robustness ratio. We conduct multiple extensive experiments on five bench mark fairness data sets using three of the most popular fairness strategies with respect to four of the most popular definitions of fairness. Our experiments empirically show that fairness methods that rely on threshold optimisation are very sensitive to noise in all the evaluated data sets, despite mostly outperforming other methods. This is in contrast to the other two methods, which are less fair for low noise scenarios but fairer for high noise ones. To the best of our knowledge, we are the first to quantitatively evaluate the robustness of fairness optimisation strategies. This can potentially can serve as a guideline in choosing the most suitable fairness strategy for various data sets.

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。

↑