发表机构
University of Electronic Science and Technology of China; Kashi Institute of Electronics and Information Industry; Intelligent Digital Media Technology（电子科技大学; 喀什电子与信息产业研究院; 智能数字媒体技术研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对标准共形预测的子组覆盖率差异及最差子组瓶颈问题，提出SGCP框架，可缩小子组覆盖率差距并保持预测集规模，且保留标准覆盖率保证。
AI 中文摘要
共形预测提供了与分布无关的覆盖率保证，使其在临床应用中极具吸引力。然而，标准共形预测仅在总体层面提供此类保证，其预测集在临床重要子组间会出现覆盖率差异。一个自然的解决方案是在预定义组内进行校准，但这可能需要访问敏感的子组属性，且易出现最差子组瓶颈：保护最困难的子组会扩大所有子组的预测集，增加决策者的认知负担。为此，我们提出随机分组共形预测（Stochastic Grouping Conformal Prediction，SGCP），这是一种用于子组可靠不确定性量化的共形框架。它学习一个随机分组映射，使每个样本能从具有相似校准行为的其他样本中获取校准信息，从而产生局部评分规律，提升子种群间的可靠性。我们证明SGCP保留了标准覆盖率保证。在合成数据集和真实基准上的实验表明，与现有基线相比，它持续缩小子组覆盖率差距，同时实现更小或相当的预测集规模。
英文摘要
Conformal prediction offers a distribution-free coverage guarantee, making it especially attractive for clinical applications. Standard conformal prediction, however, provides such guarantees only at the population level, and its prediction sets can exhibit coverage disparities across clinically important subgroups. A natural remedy is to calibrate within predefined groups. However, this can require access to sensitive subgroup attributes and is prone to a worst-group bottleneck: protecting the most difficult subgroup can inflate prediction sets for all, increasing cognitive burden on decision makers. To this end, we propose Stochastic Grouping Conformal Prediction (SGCP), a conformal framework for subgroup-reliable uncertainty quantification. It learns a stochastic grouping map that allows each sample to draw calibration information from others with similar calibration behavior, yielding a local score law that boosts reliability across subpopulations. We prove that SGCP retains the standard coverage guarantee. Experiments on synthetic and real-world benchmarks show that it consistently reduces subgroup coverage gaps while achieving smaller or comparable prediction set sizes relative to existing baselines.
Comments9 pages
