A survey on deep learning with noisy labels: How to train your model when you cannot trust on the annotations?

Date

2020

Authors

Cordeiro, F.R.
Carneiro, G.

Editors

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Conference paper

Citation

Brazilian Symposium of Computer Graphic and Image Processing, 2020, pp.9-16

Statement of Responsibility

Filipe R. Cordeiro and Gustavo Carneiro

Conference Name

Conference on Graphics, Patterns and Images (SIBGRAPI) (7 Nov 2020 - 10 Nov 2020 : virtual online)

Abstract

Noisy Labels are commonly present in data sets automatically collected from the internet, mislabeled by nonspecialist 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.

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

©2020 IEEE

License

Grant ID

Call number

Persistent link to this record