Giters

MadryLab / cifar10_challenge

A challenge to explore adversarial robustness of neural networks on CIFAR10.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

about the loss in the pgd_attack

lith0613 opened this issue · comments

commented

The loss from CW in line 36 of pgd_attack.py use a negative sign, but there is not a such sign in the original CW loss, looking forward to your help

 loss = -tf.nn.relu(correct_logit - wrong_logit + 50)
commented

The original CW attack is stated as a minimization problem. Since our PGD attack maximizes the loss, we need to modify the CW loss to accommodate for that.