NorMatch: Matching Normalizing Flows with Discriminative Classifiers for Semi-Supervised Learning

TL;DR

NorMatch introduces a novel semi-supervised learning framework combining normalizing flows with discriminative classifiers, effectively reducing bias and noise in pseudo-labels, and leveraging unlabeled data for improved performance.

Contribution

The paper proposes a new SSL framework that uses normalizing flows for uncertainty estimation and sample weighting, enhancing pseudo-label quality and model accuracy.

Findings

Achieves state-of-the-art results on multiple datasets.

Effectively reduces pseudo-label noise and bias.

Improves discriminative classifier performance through flow-based modeling.

Abstract

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.