AllMatch: Exploiting All Unlabeled Data for Semi-Supervised Learning

TL;DR

AllMatch introduces a novel semi-supervised learning method that adaptively utilizes all unlabeled data by class-specific thresholds and consistency regulation, leading to improved accuracy and full data utilization.

Contribution

The paper proposes a class-specific adaptive threshold mechanism and a binary classification consistency regulation to enhance semi-supervised learning, achieving full unlabeled data utilization.

Findings

Achieves 100% unlabeled data utilization.

Outperforms state-of-the-art methods on multiple benchmarks.

Improves pseudo-label accuracy across various settings.

Abstract

Existing semi-supervised learning algorithms adopt pseudo-labeling and consistency regulation techniques to introduce supervision signals for unlabeled samples. To overcome the inherent limitation of threshold-based pseudo-labeling, prior studies have attempted to align the confidence threshold with the evolving learning status of the model, which is estimated through the predictions made on the unlabeled data. In this paper, we further reveal that classifier weights can reflect the differentiated learning status across categories and consequently propose a class-specific adaptive threshold mechanism. Additionally, considering that even the optimal threshold scheme cannot resolve the problem of discarding unlabeled samples, a binary classification consistency regulation approach is designed to distinguish candidate classes from negative options for all unlabeled samples. By combining the above strategies, we present a novel SSL algorithm named AllMatch, which achieves improved pseudo-label accuracy and a 100% utilization ratio for the unlabeled data. We extensively evaluate our approach on multiple benchmarks, encompassing both balanced and imbalanced settings. The results demonstrate that AllMatch consistently outperforms existing state-of-the-art methods.