Unsupervised label noise modeling and loss correction
Arazo Sánchez, Eric, Ortego, DiegoORCID: 0000-0002-1011-3610, Albert, Paul, O'Connor, Noel E.ORCID: 0000-0002-4033-9135 and McGuinness, KevinORCID: 0000-0003-1336-6477
(2019)
Unsupervised label noise modeling and loss correction.
In: International Conference on Machine Learning (ICML), 10-15 June 2019, Long Beach, CA, USA.
Despite being robust to small amounts of label noise, convolutional neural networks trained with stochastic gradient methods have been shown to easily fit random labels. When there are a mixture of correct and mislabelled targets, networks
tend to fit the former before the latter. This suggests using a suitable two-component mixture model as an unsupervised generative model of sample loss values during training to allow online estimation of the probability that a sample is mislabelled. Specifically, we propose a beta mixture to estimate this probability and correct the loss by relying on the network prediction (the so-called bootstrapping loss). We further adapt mixup augmentation to drive our approach a step further. Experiments on CIFAR-10/100 and TinyImageNet demonstrate a robustness to label noise that substantially outperforms recent state-of-the-art. Source code is available at https://git.io/fjsvE and Appendix at https://arxiv.org/abs/1904.11238.