Skip to content

Conversation

jhtsing
Copy link
Contributor

@jhtsing jhtsing commented Jan 11, 2018

BIM (Kurakin et al. 2016).
Paper link: https://arxiv.org/pdf/1607.02533.pdf .
BIM based on FGSM improvement,BIM find adversarail sample by iterating multiple FGSM processes.

Paper link: https://arxiv.org/pdf/1607.02533.pdf
"""

def _apply(self, image_label, epsilons=100, steps=10):
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm curious about what's the label here. The true label? Or some fake label we made?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

image_label is the same as GradientSignAttack.
image_label(list): The image and label tuple list of one element.
The label is source label(non-target).
Target attack need new interface.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Got it. Thanks!

Copy link
Collaborator

@JiayiFeng JiayiFeng left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM. Thanks!

@jhtsing jhtsing merged commit 29e71d2 into PaddlePaddle:develop Jan 16, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants