Model and Data Agreement for Learning with Noisy Labels

TL;DR

This paper proposes a novel approach combining model and data strategies, including mean point ensemble and flip image loss evaluation, to improve learning robustness against noisy labels in deep learning datasets.

Contribution

It introduces a new method that reduces error accumulation in noisy label learning by leveraging ensemble techniques and flip image-based loss evaluation, outperforming existing methods.

Findings

Outperforms state-of-the-art noisy label learning methods on CIFAR-10, CIFAR-100, and Clothing1M.

Effectively reduces error accumulation from both model and data perspectives.

Seamlessly integrates with other noisy label learning techniques for enhanced performance.

Abstract

Learning with noisy labels is a vital topic for practical deep learning as models should be robust to noisy open-world datasets in the wild. The state-of-the-art noisy label learning approach JoCoR fails when faced with a large ratio of noisy labels. Moreover, selecting small-loss samples can also cause error accumulation as once the noisy samples are mistakenly selected as small-loss samples, they are more likely to be selected again. In this paper, we try to deal with error accumulation in noisy label learning from both model and data perspectives. We introduce mean point ensemble to utilize a more robust loss function and more information from unselected samples to reduce error accumulation from the model perspective. Furthermore, as the flip images have the same semantic meaning as the original images, we select small-loss samples according to the loss values of flip images instead of the original ones to reduce error accumulation from the data perspective. Extensive experiments on CIFAR-10, CIFAR-100, and large-scale Clothing1M show that our method outperforms state-of-the-art noisy label learning methods with different levels of label noise. Our method can also be seamlessly combined with other noisy label learning methods to further improve their performance and generalize well to other tasks. The code is available in https://github.com/zyh-uaiaaaa/MDA-noisy-label-learning.