Simplifying Neural Network Training Under Class Imbalance
Ravid Shwartz-Ziv, Micah Goldblum, Yucen Lily Li, C. Bayan Bruss, Andrew Gordon Wilson
Abstract
Real-world datasets are often highly class-imbalanced, which can adversely impact the performance of deep learning models. The majority of research on training neural networks under class imbalance has focused on specialized loss functions and sampling techniques. Notably, we demonstrate that simply tuning existing components of standard deep learning pipelines, such as the batch size, data augmentation, architecture size, pre-training, optimizer, and label smoothing, can achieve state-of-the-art performance without any specialized loss functions or samplers. We also provide key prescriptions and considerations for training under class imbalance, and an understanding of why imbalance methods succeed or fail.
BibTeX
@inproceedings{
shwartz-ziv2023simplifying,
title={Simplifying Neural Network Training Under Class Imbalance},
author={Ravid Shwartz-Ziv and Micah Goldblum and Yucen Lily Li and C. Bayan Bruss and Andrew Gordon Wilson},
booktitle={Thirty-seventh Conference on Neural Information Processing Systems},
year={2023},
url={https://openreview.net/forum?id=iGmDQn4CRj}
}