A Survey on Deep Learning with Noisy Labels: How to train your model when you cannot trust on the annotations?

TL;DR

This survey reviews techniques for training deep learning models with noisy labels, categorizing methods like robust losses, sample weighting, and meta-learning, and discusses experimental setups and results.

Contribution

It provides a comprehensive classification and analysis of existing methods to handle noisy labels in deep learning, highlighting current trends and challenges.

Findings

Robust loss functions improve model resilience to label noise.

Sample selection methods effectively reduce the impact of noisy labels.

Meta-learning approaches adaptively handle label noise during training.

Abstract

Noisy Labels are commonly present in data sets automatically collected from the internet, mislabeled by non-specialist annotators, or even specialists in a challenging task, such as in the medical field. Although deep learning models have shown significant improvements in different domains, an open issue is their ability to memorize noisy labels during training, reducing their generalization potential. As deep learning models depend on correctly labeled data sets and label correctness is difficult to guarantee, it is crucial to consider the presence of noisy labels for deep learning training. Several approaches have been proposed in the literature to improve the training of deep learning models in the presence of noisy labels. This paper presents a survey on the main techniques in literature, in which we classify the algorithm in the following groups: robust losses, sample weighting, sample selection, meta-learning, and combined approaches. We also present the commonly used experimental setup, data sets, and results of the state-of-the-art models.