精度陷阱：结构稀缺性放大算法分配中的相对不平等
The Accuracy Trap: Structural Scarcity Amplifies Relative Inequality in Algorithmic Allocation
浏览论文内容
中文总结 AI 辅助
该研究揭示了算法分配中的精度陷阱，即结构稀缺性与排名精度相互作用会放大组间相对不平等，经模拟和公共部门系统验证，仅去偏无法消除该陷阱。
中文摘要 AI 辅助
算法系统越来越多地对个体进行排名，以分配稀缺的公共资源，从儿童福利干预到癌症治疗转诊皆是如此。主流的公平框架将差异视为有偏数据或缺陷模型的属性，补救措施是校准与去偏。在结构稀缺（需求超出供给一个数量级）的情况下，分配成为配给问题，排名的统计特性与分类的统计特性截然不同。我们推导了一个缩放定律 $D \propto \exp(t \cdot \rho \cdot \Delta)$，其中由结构差距 $\Delta$ 分隔的两组间的相对差异，随稀缺诱导阈值 $t$ 与排名区分保真度 $\rho$ 的乘积而增长。稀缺性与精度呈乘法相互作用，产生指数级更大的组间差异，我们将这种动态称为精度陷阱。我们通过蒙特卡洛模拟以及加拿大儿童福利、美国癌症护理两个独立的公共部门系统验证了该精度陷阱，仅靠去偏无法消除这一陷阱。
英文摘要
Algorithmic systems increasingly rank individuals for access to scarce public resources, from child welfare interventions to cancer treatment referrals. The prevailing fairness frame treats disparity as a property of biased data or deficient models, with remedies through calibration and debiasing. Under structural scarcity, where demand exceeds supply by an order of magnitude, allocation becomes a rationing problem, and the statistical properties of ranking diverge sharply from those of classification. We derive a scaling law $D \propto \exp(t \cdot ρ\cdot Δ)$, in which relative disparity between two groups separated by a structural gap $Δ$ grows in the product of the scarcity-induced threshold $t$ and rank-discrimination fidelity $ρ$. Scarcity and accuracy interact multiplicatively, producing exponentially larger between-group disparities. We term this dynamic the Accuracy Trap. We validate this Accuracy Trap through Monte Carlo simulation and two independent public-sector systems in Canadian child welfare and U.S. cancer care. Debiasing alone cannot dissolve the trap.
