gmh14 / RobNets

[CVPR 2020] When NAS Meets Robustness: In Search of Robust Architectures against Adversarial Attacks
http://www.mit.edu/~yuzhe/robnets.html
MIT License
123 stars 15 forks source link

The natural accuracies of ResNet18 and ResNet50 in CIFAR10 #18

Open louvinci opened 3 years ago

louvinci commented 3 years ago

I am new to the field of adversarial attacks. I noticed that the natural accuracies of ResNet18 and ResNet50 in CIFAR10 are 78.38% and 79.15% respectively, which are much lower than other pre-trained models. So, what is the reason for this? Is it because of adversarial training using PGD? I also found a lack of preprocessing operations, e.g., normalization operations, prior to training the models.