AllMatch: Exploiting All Unlabeled Data for Semi-Supervised Learning
This addresses the inefficiency in semi-supervised learning for researchers and practitioners by enabling full use of unlabeled data, though it is incremental as it builds on existing pseudo-labeling and consistency techniques.
The paper tackled the problem of discarding unlabeled data in semi-supervised learning by proposing AllMatch, which uses a class-specific adaptive threshold and binary classification consistency to achieve 100% utilization of unlabeled data and improved pseudo-label accuracy, outperforming state-of-the-art methods on multiple benchmarks.
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.