Back to Search
Start Over
Simplifying Neural Network Training Under Class Imbalance
- Publication Year :
- 2023
-
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, sampling techniques, or two-stage training procedures. Notably, we demonstrate that simply tuning existing components of standard deep learning pipelines, such as the batch size, data augmentation, optimizer, and label smoothing, can achieve state-of-the-art performance without any such specialized class imbalance methods. We also provide key prescriptions and considerations for training under class imbalance, and an understanding of why imbalance methods succeed or fail.<br />Comment: NeurIPS 2023. Code available at https://github.com/ravidziv/SimplifyingImbalancedTraining
- Subjects :
- Computer Science - Machine Learning
Computer Science - Artificial Intelligence
Subjects
Details
- Database :
- arXiv
- Publication Type :
- Report
- Accession number :
- edsarx.2312.02517
- Document Type :
- Working Paper