Adaptive conformal classification with noisy labels

• 成果类型: Article
• 署名作者: Sesia, Matteo; Wang, Y. X. Rachel; Tong, Xin
• 署名单位: University of Southern California; University of Southern California; University of Sydney; University of Hong Kong
• 刊物名称: JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
• ISSN/ISSBN: 1369-7412
• DOI: 10.1093/jrsssb/qkae114
• 发表日期: 2025
• 页码: 796-815
• 关键词: predictive inference

摘要

This article develops a conformal prediction method for classification tasks that can adapt to random label contamination in the calibration sample, often leading to more informative prediction sets with stronger coverage guarantees compared to existing approaches. This is obtained through a precise characterization of the coverage inflation (or deflation) suffered by standard conformal inferences in the presence of label contamination, which is then made actionable through a new calibration algorithm. Our solution can leverage different modelling assumptions about the contamination process, while requiring no knowledge of the underlying data distribution or of the inner workings of the classification model. The empirical performance of the proposed method is demonstrated through simulations and an application to object classification with the CIFAR-10H image data set.