ICLR 2021poster62 citations
Tradeoffs in Data Augmentation: An Empirical Study
Raphael Gontijo-Lopes, Sylvia Smullin, Ekin Dogus Cubuk, Ethan Dyer
Abstract
Though data augmentation has become a standard component of deep neural network training, the underlying mechanism behind the effectiveness of these techniques remains poorly understood. In practice, augmentation policies are often chosen using heuristics of distribution shift or augmentation diversity. Inspired by these, we conduct an empirical study to quantify how data augmentation improves model generalization. We introduce two interpretable and easy-to-compute measures: Affinity and Diversity. We find that augmentation performance is predicted not by either of these alone but by jointly optimizing the two.
GeneralizationInterpretabilityUnderstanding Data Augmentation
BibTeX
@inproceedings{
gontijo-lopes2021tradeoffs,
title={Tradeoffs in Data Augmentation: An Empirical Study},
author={Raphael Gontijo-Lopes and Sylvia Smullin and Ekin Dogus Cubuk and Ethan Dyer},
booktitle={International Conference on Learning Representations},
year={2021},
url={https://openreview.net/forum?id=ZcKPWuhG6wy}
}