Heteroskedastic and Imbalanced Deep Learning with Adaptive Regularization
This addresses the challenge of distinguishing mislabeled, ambiguous, and rare examples in real-world datasets, which is an incremental advance in noise-robust deep learning.
The paper tackles the problem of heteroskedasticity and imbalance in large-scale datasets by proposing an adaptive regularization technique that regularizes higher-uncertainty, lower-density regions more heavily, resulting in significant improvements over other methods on benchmark tasks like WebVision.
Real-world large-scale datasets are heteroskedastic and imbalanced -- labels have varying levels of uncertainty and label distributions are long-tailed. Heteroskedasticity and imbalance challenge deep learning algorithms due to the difficulty of distinguishing among mislabeled, ambiguous, and rare examples. Addressing heteroskedasticity and imbalance simultaneously is under-explored. We propose a data-dependent regularization technique for heteroskedastic datasets that regularizes different regions of the input space differently. Inspired by the theoretical derivation of the optimal regularization strength in a one-dimensional nonparametric classification setting, our approach adaptively regularizes the data points in higher-uncertainty, lower-density regions more heavily. We test our method on several benchmark tasks, including a real-world heteroskedastic and imbalanced dataset, WebVision. Our experiments corroborate our theory and demonstrate a significant improvement over other methods in noise-robust deep learning.