Rebalancing with Calibrated Sub-classes (RCS): A Statistical Fusion-based Framework for Robust Imbalanced Classification across Modalities
This addresses robust imbalanced classification for real-world applications across modalities, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of class imbalance in classification by proposing the Rebalancing with Calibrated Sub-classes (RCS) framework, which uses statistical fusion to estimate minority class distributions and generate synthetic samples, resulting in consistent outperformance of baseline and state-of-the-art methods across diverse datasets.
Class imbalance, where certain classes have insufficient data, poses a critical challenge for robust classification, often biasing models toward majority classes. Distribution calibration offers a promising avenue to address this by estimating more accurate class distributions. In this work, we propose Rebalancing with Calibrated Sub-classes (RCS) - a novel distribution calibration framework for robust imbalanced classification. RCS aims to fuse statistical information from the majority and intermediate class distributions via a weighted mixture of Gaussian components to estimate minority class parameters more accurately. An encoder-decoder network is trained to preserve structural relationships in imbalanced datasets and prevent feature disentanglement. Post-training, encoder-extracted feature vectors are leveraged to generate synthetic samples guided by the calibrated distributions. This fusion-based calibration effectively mitigates overgeneralization by incorporating neighborhood distribution information rather than relying solely on majority-class statistics. Extensive experiments on diverse image, text, and tabular datasets demonstrate that RCS consistently outperforms several baseline and state-of-the-art methods, highlighting its effectiveness and broad applicability in addressing real-world imbalanced classification challenges.