Purpose of attack_pgd function

[ShengYun-Peng]

Thanks authors for the great work.
While reviewing the code, I am confused why overwriting the pgd attack again in utils.py though it was already imported from advertorch.attack.

Best,
Anthony

[nblt]

Hi Anthony,

Thanks for your question.
This pgd attack is specifically used for evaluating the robustness under PGD-50 attack (with 10 random starts), following previous implementation [1], for a fair comparison with previous methods (keeping the implementation the same).

Reference
[1] https://github.com/locuslab/robust_overfitting

[ShengYun-Peng]

Ah, I see! Thanks so much for the immediate response.
Btw, have you expanded the Sub-AT on full ImageNet?

Best,
Anthony

[nblt]

We have tried on Tiny-ImageNet, but haven't expanded on full ImageNet yet. We think that similar results can be extended to full ImageNet:)

Best,
Tao