From SMOTE to Mixup for Deep Imbalanced Classification

TL;DR

This paper enhances data augmentation techniques for deep imbalanced classification by integrating SMOTE with Mixup, introducing a margin-aware Mixup method that improves minority class generalization and achieves state-of-the-art results.

Contribution

It unifies traditional SMOTE and modern Mixup techniques, proposing a margin-aware Mixup to explicitly handle class imbalance in deep learning.

Findings

Mixup improves generalization by creating uneven class margins.

Margin-aware Mixup explicitly enhances minority class performance.

Proposed method achieves state-of-the-art results on imbalanced datasets.

Abstract

Given imbalanced data, it is hard to train a good classifier using deep learning because of the poor generalization of minority classes. Traditionally, the well-known synthetic minority oversampling technique (SMOTE) for data augmentation, a data mining approach for imbalanced learning, has been used to improve this generalization. However, it is unclear whether SMOTE also benefits deep learning. In this work, we study why the original SMOTE is insufficient for deep learning, and enhance SMOTE using soft labels. Connecting the resulting soft SMOTE with Mixup, a modern data augmentation technique, leads to a unified framework that puts traditional and modern data augmentation techniques under the same umbrella. A careful study within this framework shows that Mixup improves generalization by implicitly achieving uneven margins between majority and minority classes. We then propose a novel margin-aware Mixup technique that more explicitly achieves uneven margins. Extensive experimental results demonstrate that our proposed technique yields state-of-the-art performance on deep imbalanced classification while achieving superior performance on extremely imbalanced data. The code is open-sourced in our developed package https://github.com/ntucllab/imbalanced-DL to foster future research in this direction.