Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels
This addresses the issue of label noise hampering DNN performance for practitioners needing robust training methods, though it is incremental as it builds on existing loss functions.
The paper tackles the problem of training deep neural networks with noisy labels by proposing a set of noise-robust loss functions that generalize mean absolute error and categorical cross entropy, achieving good performance across datasets like CIFAR-10, CIFAR-100, and FASHION-MNIST with synthetic noise.
Deep neural networks (DNNs) have achieved tremendous success in a variety of applications across many disciplines. Yet, their superior performance comes with the expensive cost of requiring correctly annotated large-scale datasets. Moreover, due to DNNs' rich capacity, errors in training labels can hamper performance. To combat this problem, mean absolute error (MAE) has recently been proposed as a noise-robust alternative to the commonly-used categorical cross entropy (CCE) loss. However, as we show in this paper, MAE can perform poorly with DNNs and challenging datasets. Here, we present a theoretically grounded set of noise-robust loss functions that can be seen as a generalization of MAE and CCE. Proposed loss functions can be readily applied with any existing DNN architecture and algorithm, while yielding good performance in a wide range of noisy label scenarios. We report results from experiments conducted with CIFAR-10, CIFAR-100 and FASHION-MNIST datasets and synthetically generated noisy labels.