CVLGJan 24, 2023

Data Augmentation Alone Can Improve Adversarial Training

arXiv:2301.09879v168 citationsh-index: 64Has Code
Originality Incremental advance
AI Analysis

It addresses robust overfitting in adversarial training, which impairs generalization, but is incremental as it builds on prior findings about data augmentation.

This work tackles the problem of robust overfitting in adversarial training by demonstrating that data augmentation alone can significantly improve accuracy and robustness, achieving state-of-the-art results for data augmentations in this context.

Adversarial training suffers from the issue of robust overfitting, which seriously impairs its generalization performance. Data augmentation, which is effective at preventing overfitting in standard training, has been observed by many previous works to be ineffective in mitigating overfitting in adversarial training. This work proves that, contrary to previous findings, data augmentation alone can significantly boost accuracy and robustness in adversarial training. We find that the hardness and the diversity of data augmentation are important factors in combating robust overfitting. In general, diversity can improve both accuracy and robustness, while hardness can boost robustness at the cost of accuracy within a certain limit and degrade them both over that limit. To mitigate robust overfitting, we first propose a new crop transformation, Cropshift, which has improved diversity compared to the conventional one (Padcrop). We then propose a new data augmentation scheme, based on Cropshift, with much improved diversity and well-balanced hardness. Empirically, our augmentation method achieves the state-of-the-art accuracy and robustness for data augmentations in adversarial training. Furthermore, when combined with weight averaging it matches, or even exceeds, the performance of the best contemporary regularization methods for alleviating robust overfitting. Code is available at: https://github.com/TreeLLi/DA-Alone-Improves-AT.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes