Does Data Augmentation Benefit from Split BatchNorms
This work addresses a specific problem in computer vision for improving data augmentation techniques, but it is incremental as it builds on existing training paradigms.
The paper tackled the disparity between strongly augmented training images and clean inference images by exploring the use of an auxiliary BatchNorm for augmented data, finding that using BatchNorm parameters from weak augmentations significantly improved performance on CIFAR-10, CIFAR-100, and ImageNet benchmarks.
Data augmentation has emerged as a powerful technique for improving the performance of deep neural networks and led to state-of-the-art results in computer vision. However, state-of-the-art data augmentation strongly distorts training images, leading to a disparity between examples seen during training and inference. In this work, we explore a recently proposed training paradigm in order to correct for this disparity: using an auxiliary BatchNorm for the potentially out-of-distribution, strongly augmented images. Our experiments then focus on how to define the BatchNorm parameters that are used at evaluation. To eliminate the train-test disparity, we experiment with using the batch statistics defined by clean training images only, yet surprisingly find that this does not yield improvements in model performance. Instead, we investigate using BatchNorm parameters defined by weak augmentations and find that this method significantly improves the performance of common image classification benchmarks such as CIFAR-10, CIFAR-100, and ImageNet. We then explore a fundamental trade-off between accuracy and robustness coming from using different BatchNorm parameters, providing greater insight into the benefits of data augmentation on model performance.