Image classification with deep learning in the presence of noisy labels: A survey
作者：
Highlights：
• Label noise is a common problem in real-world datasets.
• Noise robust learning techniques are important to achieve state of the art performance.
• Many works are proposed in the literature to tackle noisy labels.
• Some works aim to estimate underlying noise structure.
• Other works try to achieve robustness without explicitly modeling the noise structure.
摘要
•Label noise is a common problem in real-world datasets.•Noise robust learning techniques are important to achieve state of the art performance.•Many works are proposed in the literature to tackle noisy labels.•Some works aim to estimate underlying noise structure.•Other works try to achieve robustness without explicitly modeling the noise structure.
论文关键词：Deep learning,Label noise,Classification with noise,Noise robust,Noise tolerant
论文评审过程：Received 20 June 2020, Revised 25 December 2020, Accepted 10 January 2021, Available online 13 January 2021, Version of Record 18 January 2021.
论文官网地址：https://doi.org/10.1016/j.knosys.2021.106771
