Adaptive conformal classification with noisy labels
This addresses the problem of robust uncertainty quantification in classification for practitioners dealing with noisy data, representing an incremental improvement in conformal prediction methods.
The paper tackled classification with noisy labels by developing adaptive conformal prediction methods that automatically adjust to label contamination, resulting in more informative prediction sets with stronger coverage guarantees compared to state-of-the-art approaches, as demonstrated through simulations and an application to CIFAR-10H.
This paper develops novel conformal prediction methods for classification tasks that can automatically adapt to random label contamination in the calibration sample, leading to more informative prediction sets with stronger coverage guarantees compared to state-of-the-art approaches. This is made possible by a precise characterization of the effective coverage inflation (or deflation) suffered by standard conformal inferences in the presence of label contamination, which is then made actionable through new calibration algorithms. Our solution is flexible and can leverage different modeling assumptions about the label contamination process, while requiring no knowledge of the underlying data distribution or of the inner workings of the machine-learning classifier. The advantages of the proposed methods are demonstrated through extensive simulations and an application to object classification with the CIFAR-10H image data set.