Add Fast Minimum-Norm Attack (FMN)

[lucascionis]

PR Type and Checklist

What kind of change does this PR introduce?

• ADD ATTACK
  • DEFAULT values of arguments should be given except model.
  • Classname should be given as CamelCase.
  • Python file name should be SMALL letters of the classname.
  • Add attack in init.py
  • Check supported_mode whether the attack supports targeted mode.
  • README.md should be updated.
  • Check WORFLOW result after PR.

Fast Minimum-Norm Attack Info

Fast Minimum-Norm Attack PyTorch implementation; works with PyTorch optimizers and schedulers and supports default and targeted mode, with L0, L1, L2 and Linf norms.

Original Code
Paper

Citation.

@article{pintor2021fast,
	title={Fast minimum-norm adversarial attacks through adaptive norm constraints},
	author={Pintor, Maura and Roli, Fabio and Brendel, Wieland and Biggio, Battista},
	journal={Advances in Neural Information Processing Systems},
	volume={34},
	year={2021}
}

[rikonaka]

Just some minor issues, waiting for maintainer approval for auto testing.

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

❌ Patch coverage is 57.59494% with 67 lines in your changes missing coverage. Please review.
✅ Project coverage is 72.74%. Comparing base (936e86d) to head (dfe601c).
⚠️ Report is 1 commits behind head on master.

Files with missing lines	Patch %	Lines
torchattacks/attacks/fmn.py	57.32%	63 Missing and 4 partials ⚠️
❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #165      +/-   ##
==========================================
- Coverage   73.37%   72.74%   -0.63%     
==========================================
  Files          44       45       +1     
  Lines        3827     3985     +158     
  Branches      578      590      +12     
==========================================
+ Hits         2808     2899      +91     
- Misses        862      925      +63     
- Partials      157      161       +4     

Files with missing lines	Coverage Δ
torchattacks/__init__.py	100.00% <100.00%> (ø)
torchattacks/attacks/fmn.py	57.32% <57.32%> (ø)

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 936e86d...dfe601c. Read the comment docs.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[maurapintor]

Hello @rikonaka, thank you for maintaining this library. Do you have any update on this PR? Is there something that we can do to speed up the merge?

[rikonaka]

Hello @rikonaka, thank you for maintaining this library. Do you have any update on this PR? Is there something that we can do to speed up the merge?

Well, all we can do now is wait 🤣. But if you have other urgent reasons for the merge, you can send an email to Harry and ask him to help merge it. As far as I know, this process may takes about 3 to 6 months. In the meantime, you can continue to check and patch the code 😁.

[maurapintor]

Hello @rikonaka, thank you for maintaining this library. Do you have any update on this PR? Is there something that we can do to speed up the merge?

Well, all we can do now is wait 🤣. But if you have other urgent reasons for the merge, you can send an email to Harry and ask him to help merge it. As far as I know, this process may takes about 3 to 6 months. In the meantime, you can continue to check and patch the code 😁.

Good, we will wait. Thank you ;)