#machine learning
Jan 2025
Noise-Adaptive Conformal Classification with Marginal Coverage
This work introduces an adaptive conformal inference method capable of efficiently handling deviations from exchangeability caused by random label noise, leading to informative prediction sets with tight marginal coverage guarantees even in those challenging scenarios.
T. Bortolotti, Y. X. Rachel Wang, Xin Tong et al.
· arXiv.org · 2 citations