Image Classification with Deep Learning in the Presence of Noisy Labels: A Survey

Image classification systems recently made a big leap with the advancement of deep neural networks. However, these systems require an excessive amount of labeled data in order to be trained properly. This is not always feasible due to several factors, such as expensiveness of labeling process or difficulty of correctly classifying data even for the experts. Because of these practical challenges, label noise is a common problem in datasets and numerous methods to train deep networks with label noise are proposed in the literature. Although deep networks are known to be relatively robust to label noise, their tendency to overfit data makes them vulnerable to memorizing even total random noise. Therefore, it is crucial to consider the existence of label noise and develop counter algorithms to fade away its negative effects to train deep neural networks efficiently. Even though an extensive survey of machine learning techniques under label noise exists, literature lacks a comprehensive survey of methodologies centered explicitly around deep learning in the presence of noisy labels. This paper aims to present these algorithms while categorizing them into one of the two subgroups: noise model based and noise model free methods. Algorithms in the first group aim to estimate the structure of the noise and use this information to avoid the negative effects of noisy labels during training. On the other hand, methods in the second group try to come up with algorithms that are inherently noise robust by using approaches like robust losses, regularizers or other learning paradigms.