Alleviating Noisy-label Effects in Image Classification via Probability Transition Matrix

TL;DR

This paper introduces a noise ignoring block (NIB) with a probability transition matrix and inter-class correlation loss to better distinguish hard samples from mislabeled ones, improving image classification accuracy under noisy labels.

Contribution

The proposed NIB module, combining a probability transition matrix and IC loss, effectively separates hard samples from noisy labels, enhancing robustness of training methods.

Findings

NIB improves classification accuracy on CIFAR-10 and ISIC 2019 datasets.

NIB consistently outperforms existing robust training methods.

The IC loss effectively distinguishes hard samples from mislabeled data.

Abstract

Deep-learning-based image classification frameworks often suffer from the noisy label problem caused by the inter-observer variation. Recent studies employed learning-to-learn paradigms (e.g., Co-teaching and JoCoR) to filter the samples with noisy labels from the training set. However, most of them use a simple cross-entropy loss as the criterion for noisy label identification. The hard samples, which are beneficial for classifier learning, are often mistakenly treated as noises in such a setting since both the hard samples and ones with noisy labels lead to a relatively larger loss value than the easy cases. In this paper, we propose a plugin module, namely noise ignoring block (NIB), consisting of a probability transition matrix and an inter-class correlation (IC) loss, to separate the hard samples from the mislabeled ones, and further boost the accuracy of image classification network trained with noisy labels. Concretely, our IC loss is calculated as Kullback-Leibler divergence between the network prediction and the accumulative soft label generated by the probability transition matrix. Such that, with the lower value of IC loss, the hard cases can be easily distinguished from mislabeled cases. Extensive experiments are conducted on natural and medical image datasets (CIFAR-10 and ISIC 2019). The experimental results show that our NIB module consistently improves the performances of the state-of-the-art robust training methods.