How to train your ViT? Data, Augmentation, and Regularization in Vision Transformers
This addresses the problem of data efficiency for researchers and practitioners using Vision Transformers, showing incremental improvements in training strategies.
The study systematically investigates how training data, augmentation, regularization, model size, and compute interact in Vision Transformers, finding that increased compute and AugReg can match performance of models trained on 10x more data, with ViTs on ImageNet-21k matching or outperforming those on JFT-300M.
Vision Transformers (ViT) have been shown to attain highly competitive performance for a wide range of vision applications, such as image classification, object detection and semantic image segmentation. In comparison to convolutional neural networks, the Vision Transformer's weaker inductive bias is generally found to cause an increased reliance on model regularization or data augmentation ("AugReg" for short) when training on smaller training datasets. We conduct a systematic empirical study in order to better understand the interplay between the amount of training data, AugReg, model size and compute budget. As one result of this study we find that the combination of increased compute and AugReg can yield models with the same performance as models trained on an order of magnitude more training data: we train ViT models of various sizes on the public ImageNet-21k dataset which either match or outperform their counterparts trained on the larger, but not publicly available JFT-300M dataset.