Skip to yearly menu bar Skip to main content


Poster

Heteroskedastic and Imbalanced Deep Learning with Adaptive Regularization

Kaidi Cao · Yining Chen · Junwei Lu · Nikos Arechiga · Adrien Gaidon · Tengyu Ma

Keywords: [ deep learning ] [ noise robust learning ] [ imbalanced learning ]


Abstract:

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.

Chat is not available.