Revisiting Confidence Estimation: Towards Reliable Failure Prediction

arXiv:2403.02886v1 Announce Type: cross
Abstract: Reliable confidence estimation is a challenging yet fundamental requirement in many risk-sensitive applications. However, modern deep neural networks are often overconfident for their incorrect predictions, i.e., misclassified samples from known classes, and out-of-distribution (OOD) samples from unknown classes. In recent years, many confidence calibration and OOD detection methods have been developed. In this paper, we find a general, widely existing but actually-neglected phenomenon that most confidence estimation methods are harmful for detecting misclassification errors. We …

arxiv confidence cs.cv cs.lg failure failure prediction prediction type