Uncertainty Based Detection and Relabeling of Noisy Image Labels
This addresses the issue of label noise in computer vision for practitioners, but it is incremental as it builds on existing uncertainty-based techniques.
The paper tackles the problem of noisy labels in training images for deep neural networks, which harms generalization, by proposing a method to detect and relabel noisy data using predictive uncertainty distributions and training behavior, achieving promising performance on CIFAR-10 and CIFAR-100.
Deep neural networks (DNNs) are powerful tools in computer vision tasks. However, in many realistic scenarios label noise is prevalent in the training images, and overfitting to these noisy labels can significantly harm the generalization performance of DNNs. We propose a novel technique to identify data with noisy labels based on the different distributions of the predictive uncertainties from a DNN over the clean and noisy data. Additionally, the behavior of the uncertainty over the course of training helps to identify the network weights which best can be used to relabel the noisy labels. Data with noisy labels can therefore be cleaned in an iterative process. Our proposed method can be easily implemented, and shows promising performance on the task of noisy label detection on CIFAR-10 and CIFAR-100.