Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Leveraging Label Proportion Prior for Class-Imbalanced Semi-Supervised Learning

Kohki Akiba, Shinnosuke Matsuo, Shota Harada, Ryoma Bise

TL;DR

This work introduces Proportion Loss from learning from label proportions (LLP) into SSL as a regularization term, mitigating bias across both majority and minority classes, and forms a stochastic variant that accounts for fluctuations in mini-batch composition.

Abstract

Semi-supervised learning (SSL) often suffers under class imbalance, where pseudo-labeling amplifies majority bias and suppresses minority performance. We address this issue with a lightweight framework that, to our knowledge, is the first to introduce Proportion Loss from learning from label proportions (LLP) into SSL as a regularization term. Proportion Loss aligns model predictions with the global class distribution, mitigating bias across both majority and minority classes. To further stabilize training, we formulate a stochastic variant that accounts for fluctuations in mini-batch composition. Experiments on the Long-tailed CIFAR-10 benchmark show that integrating Proportion Loss into FixMatch and ReMixMatch consistently improves performance over the baselines across imbalance severities and label ratios, and achieves competitive or superior results compared to existing CISSL methods, particularly under scarce-label conditions.

Leveraging Label Proportion Prior for Class-Imbalanced Semi-Supervised Learning

TL;DR

This work introduces Proportion Loss from learning from label proportions (LLP) into SSL as a regularization term, mitigating bias across both majority and minority classes, and forms a stochastic variant that accounts for fluctuations in mini-batch composition.

Abstract

Semi-supervised learning (SSL) often suffers under class imbalance, where pseudo-labeling amplifies majority bias and suppresses minority performance. We address this issue with a lightweight framework that, to our knowledge, is the first to introduce Proportion Loss from learning from label proportions (LLP) into SSL as a regularization term. Proportion Loss aligns model predictions with the global class distribution, mitigating bias across both majority and minority classes. To further stabilize training, we formulate a stochastic variant that accounts for fluctuations in mini-batch composition. Experiments on the Long-tailed CIFAR-10 benchmark show that integrating Proportion Loss into FixMatch and ReMixMatch consistently improves performance over the baselines across imbalance severities and label ratios, and achieves competitive or superior results compared to existing CISSL methods, particularly under scarce-label conditions.
Paper Structure (13 sections, 4 equations, 4 figures, 1 table)

This paper contains 13 sections, 4 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related work
  3. Proportion-Regularized Semi-Supervised Learning
  4. Problem Setting
  5. Proportion Regularization for SSL
  6. Regularization with Proportion Loss
  7. Proportion Perturbation via Hypergeometric Sampling
  8. Experiment
  9. Experimental Setup
  10. Comparison of Accuracy
  11. Estimated Proportion after Training
  12. Analysis of Pseudo-Label Selection
  13. Conclusion

Figures (4)

  • Figure 1: Our approach: Regularization with the label proportion prior for semi-supervised learning (SSL).
  • Figure 2: Overview of the proposed method.
  • Figure 3: Comparison of estimated output proportions after training. Red indicates overestimation and blue indicates underestimation across classes (Class 1 = major, Classes 7–9 = minor).
  • Figure 4: Recall of pseudo-labels for the most major class (left) and most minor class (right) during training.