Papers

arXiv 论文:公平策略在部署后可能比 Pareto 最优策略成本更低

The Delayed Societal Cost of Unfairness

Why we picked it

这篇 arXiv 论文用动态人群模型证明,静态数据上看似牺牲性能的公平策略,部署后可能反而降低总成本。做信用评分、招聘筛选这类长期决策系统的研究者和工程师该看看,评估方式可能要改。

arXiv 论文《The Delayed Societal Cost of Unfairness》研究决策政策对人群结构的动态影响。传统公平机器学习在静态数据上评估,公平与性能常被视为此消彼长的权衡。

Body

论文设定资格分数会随二值决策及其结果演化。结果表明:在 Rawlsian 公平(不降低任何群体的接受率)下,当接受处于 DM 盈亏平衡点的弱势群体使其对 DM 的成本低于拒绝时,对公平赋予足够高的权重即可降低社会总成本;在 egalitarian 公平下,还需弱势群体上 DM 损失的下降超过优势群体上的上升。两种公平概念下均发现一个反直觉现象:当 DM 损失基于准确率时,部署后人群的改善可能源于资格下降,即 DM 的收益不等于 DS 的更好结果。

结论是政策诱导的人群动态可以推翻表面上存在的性能-公平权衡,更公平的策略反而能降低决策的延迟社会成本,这对贷款、招聘等长期决策场景的评估方法有直接提示。

Original · arXiv cs.LG

The Delayed Societal Cost of Unfairness

Fair machine learning typically evaluates the societal costs of a decision policy -- the loss incurred by the decision-maker (DM) and the unfair distribution of costs among demographic groups of decision-subjects (DS) -- on static data, where reducing one often comes at the expense of the other. Yet decisions change the populations on which they act, and prior work has shown that fairness interventions designed on static data may fail to benefit, or even harm, the disadvantaged group over time. We investigate a complementary question: When can a policy that is Pareto-optimal on pre-deployment data be outperformed by a fairer policy once each is evaluated on the population it induces, reducing costs for one stakeholder without increasing them for the other? In a setting where qualification scores evolve in response to binary decisions and their outcomes, we show that under Rawlsian fairness, which never lowers any group's acceptance rate, placing sufficiently high weight on fairness reduces societal costs whenever accepting disadvantaged DS at the DM's break-even point leaves them in a state less costly to the DM than rejecting them. Under egalitarian fairness, which may lower the advantaged group's acceptance rate, this additionally requires the reduction in DM loss on the disadvantaged group to outweigh the increase on the advantaged group. Under both fairness notions, however, when DM loss is accuracy-based, improvements in the post-deployment population may arise from declining DS qualification, revealing that gains for the DM do not necessarily translate into better outcomes for DS. Our results show that policy-induced population dynamics can overturn apparent performance-fairness trade-offs, allowing fairer policies to reduce the delayed societal costs of decision-making.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Papers" desk. Key entities and technical topics under continuous tracking include Rawlsian fairness, Pareto optimality, algorithmic fairness.

Source attribution: curated from arXiv cs.LG (@Giovani Valdrighi, Isabel Valera, Marcos Medeiros Raimundo).