Untapped Potential of Data Augmentation: A Domain Generalization Viewpoint
arXiv:2007.04662
Abstract
Data augmentation is a popular pre-processing trick to improve generalization accuracy. It is believed that by processing augmented inputs in tandem with the original ones, the model learns a more robust set of features which are shared between the original and augmented counterparts. However, we show that is not the case even for the best augmentation technique. In this work, we take a Domain Generalization viewpoint of augmentation based methods. This new perspective allowed for probing overfitting and delineating avenues for improvement. Our exploration with the state-of-art augmentation method provides evidence that the learned representations are not as robust even towards distortions used during training. This suggests evidence for the untapped potential of augmented examples.
6 pages, ICML 2020 Workshop on Uncertainty and Ro-bustness in Deep Learning
References in corpus (8)
- ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness
- AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
- Data Augmentation using Random Image Cropping and Patching for Deep CNNs
- Generalizing Across Domains via Cross-Gradient Training
- Learning Robust Global Representations by Penalizing Local Predictive Power
- Examining the Impact of Blur on Recognition by Convolutional Networks
- Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation
- Efficient Domain Generalization via Common-Specific Low-Rank Decomposition