发表机构
University of Florida; Wichita State University; Nanchang Hangkong University（佛罗里达大学; 威奇托州立大学; 南昌航空大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出EcoTrust框架，结合风险估计与置信度规则，在环境AI案例中比较自动操作与审查的成本，发现丰富风险信号未必优于标定置信度。
AI 中文摘要
为冻结的分类器选择延迟决策策略，不仅需要对不确定案例进行排序：置信度可能未被正确标定，错误具有不等的成本，审查者可能出错，且部署数据可能超出标定支持范围。我们通过EcoTrust框架研究这些交互作用，该框架是一个后验框架，使用六组错误风险估计器、类别不对称成本、审查者准确性以及可选的支持门控，将自动操作与审查进行比较。在哥伦比亚河热应力测试平台上，学习到的估计器将错误排序的接受者操作特征曲线下面积从0.869提高到0.889，但Chow的置信度规则具有更低的分布内成本（每天0.416对比0.567）。在12个现成后端中，学习到的风险以及标定匹配的、类别感知的置信度估计器在六个后端上各自优于原始Chow；配对的年份块自助法无法解决其平均成本差异。在转移到十个河流站点时，门控标记每个案例并成为始终审查的回退机制，仅在审查完美且不受限制时，在八个站点上达到最低成本。这些结果刻画了一个受控任务上的决策边界：更丰富的风险信号并不能可靠地优于标定后的置信度，且检测到的外推并不暗示可转移的案例级排序。
英文摘要
Choosing a deferral policy for a frozen classifier requires more than ranking uncertain cases: confidence may be miscalibrated, errors have unequal costs, reviewers can err, and deployment data can leave calibration support. We study these interactions through EcoTrust, a post-hoc framework that compares automatic action with review using a six-group error-risk estimator, class-asymmetric costs, reviewer accuracy, and an optional support gate. On a Columbia River thermal-stress testbed, the learned estimator improves error-ranking area under the receiver operating characteristic curve from 0.869 to 0.889, but Chow's confidence rule has lower in-distribution cost (0.416 versus 0.567 per day). Across 12 off-the-shelf backends, learned risk and a calibration-matched, class-aware confidence estimator each beat raw Chow on six; a paired year-block bootstrap does not resolve their mean cost difference. In transfer to ten river stations, the gate flags every case and becomes an always-review fallback, attaining the lowest cost on eight stations only when review is perfect and unconstrained. These results characterize decision boundaries on one controlled task: richer risk signals do not reliably improve on calibrated confidence, and detected extrapolation does not imply transferable case-level ranking.
