NorMatch: Matching Normalizing Flows with Discriminative Classifiers for Semi-Supervised Learning
This work addresses semi-supervised learning challenges for researchers and practitioners by improving pseudo-label quality, though it appears incremental as it builds on existing SSL methods.
The paper tackles the problem of confirmation bias and noise in pseudo-labels for semi-supervised learning by introducing NorMatch, a framework that uses normalizing flows for uncertainty estimation and distribution modeling, achieving state-of-the-art performance on several datasets.
Semi-Supervised Learning (SSL) aims to learn a model using a tiny labeled set and massive amounts of unlabeled data. To better exploit the unlabeled data the latest SSL methods use pseudo-labels predicted from a single discriminative classifier. However, the generated pseudo-labels are inevitably linked to inherent confirmation bias and noise which greatly affects the model performance. In this work we introduce a new framework for SSL named NorMatch. Firstly, we introduce a new uncertainty estimation scheme based on normalizing flows, as an auxiliary classifier, to enforce highly certain pseudo-labels yielding a boost of the discriminative classifiers. Secondly, we introduce a threshold-free sample weighting strategy to exploit better both high and low confidence pseudo-labels. Furthermore, we utilize normalizing flows to model, in an unsupervised fashion, the distribution of unlabeled data. This modelling assumption can further improve the performance of generative classifiers via unlabeled data, and thus, implicitly contributing to training a better discriminative classifier. We demonstrate, through numerical and visual results, that NorMatch achieves state-of-the-art performance on several datasets.