Advancements in Active Learning: Strategies for Imbalanced Class Settings

Active learning (AL) is a machine learning technique that selects the most informative samples from a large pool of unlabeled data for annotation, thus reducing the labeling cost and improving the learning performance. However, conventional AL approaches often neglect the intricate issue of class imbalance, where certain classes are either overrepresented or underrepresented in the dataset distribution. This disparity can introduce bias in sampling and compromise the overall generalization ability of the classifier. In this work, we introduce a novel threshold-based strategy for AL designed to navigate the challenges of class imbalance. This strategy dynamically adjusts to the degree of class imbalance, ensuring the selection of samples that are both informative and well-representative of minority classes. Our approach is rigorously tested on a variety of imbalanced datasets and benchmarked against state-of-the-art AL methods. Empirical results demonstrate that our proposed method significantly enhances classifier performance, especially in scenarios characterized by imbalanced class labels.