Improved Mix-up with KL-Entropy for Learning From Noisy Labels
This work addresses the challenge of noisy labels in image classification, which is a common issue in real-world datasets, but it appears incremental as it builds on existing mix-up and entropy-based methods.
The paper tackles the problem of training deep neural networks on datasets with noisy labels, which can cause overfitting and performance degradation, by proposing an improved joint optimization framework that mixes mix-up entropy and KL entropy as the loss function, achieving advantageous performance on CIFAR-10 and Clothing1M datasets compared to state-of-the-art methods.
Despite the deep neural networks (DNN) has achieved excellent performance in image classification researches, the training of DNNs needs a large of clean data with accurate annotations. The collect of a dataset is easy, but it is difficult to annotate the collecting data. On the websites, there exist a lot of image data which contains inaccurate annotations, but training on these datasets may make networks easier to over-fit the noisy labels and cause performance degradation. In this work, we propose an improved joint optimization framework, which mixed the mix-up entropy and Kullback-Leibler (KL) entropy as the loss function. The new loss function can give the better fine-tuning after the framework updates both the label annotations. We conduct experiments on CIFAR-10 dataset and Clothing1M dataset. The result shows the advantageous performance of our approach compared with other state-of-the-art methods.