Drawing Multiple Augmentation Samples Per Image During Training Efficiently Decreases Test Error

Abstract

In computer vision, it is standard practice to draw a single sample from the data augmentation procedure for each unique image in the mini-batch. However recent work has suggested drawing multiple samples can achieve higher test accuracies. In this work, we provide a detailed empirical evaluation of how the number of augmentation samples per unique image influences model performance on held out data when training deep ResNets. We demonstrate drawing multiple samples per image consistently enhances the test accuracy achieved for both small and large batch training. Crucially, this benefit arises even if different numbers of augmentations per image perform the same number of parameter updates and gradient evaluations (requiring the same total compute). Although prior work has found variance in the gradient estimate arising from subsampling the dataset has an implicit regularization benefit, our experiments suggest variance which arises from the data augmentation process harms generalization. We apply these insights to the highly performant NFNet-F5, achieving 86.8$\%$ top-1 w/o extra data on ImageNet.

References in corpus (6)

• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• High-Performance Large-Scale Image Recognition Without Normalization
• Augment your batch: better training with larger batches
• On the Origin of Implicit Regularization in Stochastic Gradient Descent
• On the Generalization Benefit of Noise in Stochastic Gradient Descent
• The Effect of Network Width on Stochastic Gradient Descent and Generalization: an Empirical Study

Cited by in corpus (2)

• VOLO: Vision Outlooker for Visual Recognition
• Data augmentation in Bayesian neural networks and the cold posterior effect