Do not trust what you trust: Miscalibration in Semi-supervised Learning

arXiv:2403.15567v1 Announce Type: new
Abstract: State-of-the-art semi-supervised learning (SSL) approaches rely on highly confident predictions to serve as pseudo-labels that guide the training on unlabeled samples. An inherent drawback of this strategy stems from the quality of the uncertainty estimates, as pseudo-labels are filtered only based on their degree of uncertainty, regardless of the correctness of their predictions. Thus, assessing and enhancing the uncertainty of network predictions is of paramount importance in the pseudo-labeling process. In this work, we empirically …

abstract art arxiv cs.cv cs.lg guide labels predictions quality samples semi-supervised semi-supervised learning serve ssl state strategy supervised learning training trust type uncertainty