Class-wise Flooding Regularization for Imbalanced Image Classification

TL;DR

This paper introduces class-wise flooding regularization, a novel method that adjusts regularization levels per class to improve minority class recognition in imbalanced image datasets.

Contribution

It extends flooding regularization to class-specific levels, effectively balancing learning between majority and minority classes in imbalanced datasets.

Findings

Improves minority class classification performance.

Enhances overall generalization in imbalanced image classification.

Outperforms conventional flooding methods.

Abstract

The purpose of training neural networks is to achieve high generalization performance on unseen inputs. However, when trained on imbalanced datasets, a model's prediction tends to favor majority classes over minority classes, leading to significant degradation in the recognition performance of minority classes. To address this issue, we propose class-wise flooding regularization, an extension of flooding regularization applied at the class level. Flooding is a regularization technique that mitigates overfitting by preventing the training loss from falling below a predefined threshold, known as the flooding level, thereby discouraging memorization. Our proposed method assigns a class-specific flooding level based on class frequencies. By doing so, it suppresses overfitting in majority classes while allowing sufficient learning for minority classes. We validate our approach on imbalanced image classification. Compared to conventional flooding regularizations, our method improves the classification performance of minority classes and achieves better overall generalization.