arXiv cs.AI· Meihui Zhong, Wenxin Tai, Ting Zhong, Fan Zhou·· 9 天前AI 评分22
SGCP：面向子群可靠性的随机分组共形预测框架
Stochastic Grouping Conformal Prediction for Effective Subgroup Reliability
AI 导读
研究者提出随机分组共形预测（SGCP），一种面向子群可靠性的不确定性量化框架，通过学习随机分组映射让每个样本从校准行为相似的样本中获取校准信息，形成局部分数分布。SGCP 被证明保留标准覆盖率保证，在合成与真实基准上持续缩小各子群覆盖率差距，同时预测集大小较现有基线更小或相当。
正文 · 原文
Abstract: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.
| Comments: | 9 pages |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.11957 [cs.LG] |
| (or arXiv:2610.11957v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11957 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Meihui Zhong [view email]
[v1]
Thu, 8 Oct 2026 13:41:10 UTC (1,535 KB)
来源：arXiv cs.AI · arxiv.org