USE: Uncertainty Structure Estimation for Robust Semi-Supervised Learning
This addresses the bottleneck of unlabeled data quality for SSL practitioners, offering a practical solution for realistic mixed-distribution environments, though it is incremental as it reframes an existing issue rather than introducing a new paradigm.
The paper tackles the problem of unreliable semi-supervised learning (SSL) due to contaminated unlabeled data by introducing Uncertainty Structure Estimation (USE), a lightweight preprocessing method that assesses and curates unlabeled data quality, resulting in improved accuracy and robustness in experiments on CIFAR-100 and Yelp Review datasets.
In this study, a novel idea, Uncertainty Structure Estimation (USE), a lightweight, algorithm-agnostic procedure that emphasizes the often-overlooked role of unlabeled data quality is introduced for Semi-supervised learning (SSL). SSL has achieved impressive progress, but its reliability in deployment is limited by the quality of the unlabeled pool. In practice, unlabeled data are almost always contaminated by out-of-distribution (OOD) samples, where both near-OOD and far-OOD can negatively affect performance in different ways. We argue that the bottleneck does not lie in algorithmic design, but rather in the absence of principled mechanisms to assess and curate the quality of unlabeled data. The proposed USE trains a proxy model on the labeled set to compute entropy scores for unlabeled samples, and then derives a threshold, via statistical comparison against a reference distribution, that separates informative (structured) from uninformative (structureless) samples. This enables assessment as a preprocessing step, removing uninformative or harmful unlabeled data before SSL training begins. Through extensive experiments on imaging (CIFAR-100) and NLP (Yelp Review) data, it is evident that USE consistently improves accuracy and robustness under varying levels of OOD contamination. Thus, it can be concluded that the proposed approach reframes unlabeled data quality control as a structural assessment problem, and considers it as a necessary component for reliable and efficient SSL in realistic mixed-distribution environments.