WeMix: How to Better Utilize Data Augmentation
This work addresses a fundamental limitation in data augmentation for deep learning practitioners, offering a novel solution to enhance generalization, though it is incremental in building on existing augmentation methods.
The paper tackles the problem of data bias in conventional data augmentation, which limits performance, and proposes two algorithms, AugDrop and MixLoss, to correct this bias, leading to improved effectiveness as validated by empirical studies.
Data augmentation is a widely used training trick in deep learning to improve the network generalization ability. Despite many encouraging results, several recent studies did point out limitations of the conventional data augmentation scheme in certain scenarios, calling for a better theoretical understanding of data augmentation. In this work, we develop a comprehensive analysis that reveals pros and cons of data augmentation. The main limitation of data augmentation arises from the data bias, i.e. the augmented data distribution can be quite different from the original one. This data bias leads to a suboptimal performance of existing data augmentation methods. To this end, we develop two novel algorithms, termed "AugDrop" and "MixLoss", to correct the data bias in the data augmentation. Our theoretical analysis shows that both algorithms are guaranteed to improve the effect of data augmentation through the bias correction, which is further validated by our empirical studies. Finally, we propose a generic algorithm "WeMix" by combining AugDrop and MixLoss, whose effectiveness is observed from extensive empirical evaluations.