A Implementation details Image augmentations BYOL uses the same set of image augmentations as in SimCLR

Additional details on the image augmentations are in Appendix C. Architecture Table 6: Parameters used to generate image augmentations. This method achieves better performance both in top-1 and top-5 accuracy. We report in Table 7 the top-1 and top-5 accuracy on ImageNet using this modified protocol. At training time, we apply spatial augmentations, i.e., random Note that large networks are facing overfitting problems. Top-1 and top-5 accuracies are reported in %. Results are presented in Figure 6.