Robust Asymmetric Loss for Multi-Label Long-Tailed Learning

TL;DR

This paper introduces a robust asymmetric loss function tailored for multi-label, long-tailed medical image classification, effectively handling class imbalance and multi-label issues with reduced overfitting risk.

Contribution

The paper proposes a novel robust asymmetric loss with Hill loss regularization that addresses long-tailed and multi-label classification challenges simultaneously, improving medical image classification performance.

Findings

Outperforms existing methods on long-tailed multi-label medical datasets.

Achieves Top-5 results on the CXR-LT dataset in ICCV CVAMD 2023.

Provides a generic, efficient loss function applicable to various medical image tasks.

Abstract

In real medical data, training samples typically show long-tailed distributions with multiple labels. Class distribution of the medical data has a long-tailed shape, in which the incidence of different diseases is quite varied, and at the same time, it is not unusual for images taken from symptomatic patients to be multi-label diseases. Therefore, in this paper, we concurrently address these two issues by putting forth a robust asymmetric loss on the polynomial function. Since our loss tackles both long-tailed and multi-label classification problems simultaneously, it leads to a complex design of the loss function with a large number of hyper-parameters. Although a model can be highly fine-tuned due to a large number of hyper-parameters, it is difficult to optimize all hyper-parameters at the same time, and there might be a risk of overfitting a model. Therefore, we regularize the loss function using the Hill loss approach, which is beneficial to be less sensitive against the numerous hyper-parameters so that it reduces the risk of overfitting the model. For this reason, the proposed loss is a generic method that can be applied to most medical image classification tasks and does not make the training process more time-consuming. We demonstrate that the proposed robust asymmetric loss performs favorably against the long-tailed with multi-label medical image classification in addition to the various long-tailed single-label datasets. Notably, our method achieves Top-5 results on the CXR-LT dataset of the ICCV CVAMD 2023 competition. We opensource our implementation of the robust asymmetric loss in the public repository: https://github.com/kalelpark/RAL.