Improving Transformer Optimization Through Better Initialization,ICML20

code

InstaBoost: Boosting Instance Segmentation via Probability Map Guided Copy-Pasting,ICCV19

jittering guided by appearance consistency map

Mixup Inference: Better Exploiting Mixup to Defend Adversarial Attacks,ICLR20

mixup: Beyond Empirical Risk Minimization,ICLR18

We use mixup and ERM to train several state-of-the-art ImageNet-2012 classification models, and report both top-1 and top-5 error rates in Table 1.

**[Manifold Mixup: Better Representations by Interpolating Hidden States](https://arxiv.org/pdf/1806.05236.pdf)**

Hide and Seek <!–

https://zhuanlan.zhihu.com/p/

–> Cutout

CutMix

GridMask