Learning Label Refinement and Threshold Adjustment for Imbalanced Semi-Supervised Learning
This addresses a specific bottleneck in SSL for imbalanced data, offering an incremental improvement to enhance pseudo-labeling techniques.
The paper tackles the problem of semi-supervised learning (SSL) struggling with imbalanced training data, where pseudo-labels become biased towards majority classes, by introducing SEVAL, which learns refinement and thresholding parameters from validation data to improve pseudo-label quality, resulting in surpassing state-of-the-art SSL methods in various imbalanced scenarios.
Semi-supervised learning (SSL) algorithms struggle to perform well when exposed to imbalanced training data. In this scenario, the generated pseudo-labels can exhibit a bias towards the majority class, and models that employ these pseudo-labels can further amplify this bias. Here we investigate pseudo-labeling strategies for imbalanced SSL including pseudo-label refinement and threshold adjustment, through the lens of statistical analysis. We find that existing SSL algorithms which generate pseudo-labels using heuristic strategies or uncalibrated model confidence are unreliable when imbalanced class distributions bias pseudo-labels. To address this, we introduce SEmi-supervised learning with pseudo-label optimization based on VALidation data (SEVAL) to enhance the quality of pseudo-labelling for imbalanced SSL. We propose to learn refinement and thresholding parameters from a partition of the training dataset in a class-balanced way. SEVAL adapts to specific tasks with improved pseudo-labels accuracy and ensures pseudo-labels correctness on a per-class basis. Our experiments show that SEVAL surpasses state-of-the-art SSL methods, delivering more accurate and effective pseudo-labels in various imbalanced SSL situations. SEVAL, with its simplicity and flexibility, can enhance various SSL techniques effectively. The code is publicly available (https://github.com/ZerojumpLine/SEVAL).