A Light Recipe to Train Robust Vision Transformers

Computer Science

• Vision Transformer 100%
• Adversarial Machine Learning 83%
• Data Augmentation 33%
• Underlying Architecture 16%
• Convolutional Neural Network 16%
• Machine Learning Model 16%
• Convolution 16%
• Adversarial Robustness 16%

Keyphrases

• Strong Augmentation 16%
• ImageNet-1K 16%