Can I read ConvNets Match Vision Transformers at Scale on EtoBox?

ConvNets Match Vision Transformers at Scale by Smith, Samuel L.; Brock, Andrew; Berrada, Leonard; De, Soham is a scholarly article available to read on EtoBox.

What is ConvNets Match Vision Transformers at Scale about?

Many researchers believe that ConvNets perform well on small or moderately sized datasets, but are not competitive with Vision Transformers when given access to datasets on the web-scale. We challenge this belief by evaluating a performant ConvNet architecture pre-trained on JFT-4B, a large labelled dataset of images often used for training foundation models. We consider pre-training compute budgets between 0.4k and 110k TPU-v4 core compute hours, and train a series of networks of increasing depth and width from the NFNet model family. We observe a log-log scaling law between held out loss and compute budget. After fine-tuning on ImageNet, NFNets match the reported performance of Vision Transformers with comparable compute budgets. Our strongest fine-tuned model achieves a Top-1 accuracy of 90.4%.

Author

Smith, Samuel L.; Brock, Andrew; Berrada, Leonard; De, Soham

Published

2023

Language

EN