fra31 auto-attack issues

fra31 / auto-attack

Code relative to "Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks"

https://arxiv.org/abs/2003.01690

MIT License

626 stars 111 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

update .gitignore

#107 userElaina opened 2 months ago
0
About `n_iter`, to align with others

#106 userElaina closed 1 month ago
4
Fixed evaluation of models with random defenses

#105 Buntender opened 6 months ago
2
Fixed random escape from attack problem of random evaluation

#104 LYMDLUT closed 6 months ago
0
normalization with deepfool fmodel

#103 jS5t3r closed 9 months ago
4
Softmax probabilities instead of logits

#102 fugokidi closed 1 year ago
3
fix hard coding

#101 s-kumano closed 1 year ago
1
Can AutoAttack be used to dense prediction task?

#100 zhanglijun95 opened 1 year ago
2
TF1: How do labels get modified by batch_datapoint_idcs?

#99 giorgiopiras closed 1 year ago
8
Create attack state

#98 dedeswim closed 1 year ago
8
Using the tool for general interval of input region

#97 shubhamugare closed 1 year ago
3
hotfix for #95

#96 CNOCycle closed 1 year ago
1
Invalid configuration for square attack

#95 CNOCycle closed 1 year ago
2
Parallelized computing

#94 Bai-YT closed 1 year ago
5
Minor improvement for APGD

#93 CNOCycle closed 1 year ago
1
fix check for tf models

#92 fra31 closed 1 year ago
1
Bug Fix | FAB attack targetted version inable to reach

#91 HengyueL closed 2 years ago
2
remove unnecessary var and loop

#90 s-kumano closed 2 years ago
2
Unable to use in TF2

#89 Shubhajit1988 opened 2 years ago
5
Unable to use Python debugger

#88 yuezuegu opened 2 years ago
4
gradient computation issue

#87 CNOCycle closed 2 years ago
4
max Linf perturbation is larger than the epsilon

#86 XiaoYaoYouUSTC closed 11 months ago
3
ensemble attack

#85 Jialiang14 opened 2 years ago
3
"The connection to the C10d store has failed" on distributed evaluation

#84 chawins opened 2 years ago
2
Problem that forward output is tuple rather than tensor

#83 momo1986 opened 2 years ago
1
handle corner case in square and update readme

#82 fra31 closed 2 years ago
0
A Bug in Square Attack

#81 GuanlinLee closed 2 years ago
4
throw VE when attacks to run is specified but version is not 'custom'

#80 niklasnolte closed 2 years ago
3
4 classes model; targeted attacks are different

#79 jS5t3r opened 2 years ago
2
Import error for pytorch version 1.10.0

#78 henrikju closed 2 years ago
3
Forspectraldefense

#77 jS5t3r closed 2 years ago
0
eps 8./255. works fine 4./255. does not work fine.

#76 jS5t3r closed 2 years ago
12
add checks about number of classes

#75 fra31 closed 2 years ago
0
Question about variation in reported (clean & robust accuracy) metrics

#74 choltz95 closed 2 years ago
2
Regarding the checks recently added to AutoAttack

#73 zangobot closed 2 years ago
4
update readme with checks and frozen evaluations

#72 fra31 closed 2 years ago
0
Add checks

#71 fra31 closed 2 years ago
0
The DLR loss is undefined for classification problems with fewer than 4 classes

#70 peck94 opened 2 years ago
1
Add FSGM_APR_SP

#69 iCGY96 closed 2 years ago
4
Update AutoAttack Framework to pytorch 1.9

#68 jS5t3r closed 2 years ago
2
return y_adv

#67 MariaMsu closed 2 years ago
4
replace '.view()' with '.reshape()'

#66 MariaMsu closed 3 years ago
0
remove unnecessary files

#65 CNOCycle closed 3 years ago
0
Adversarials are equals to originals

#64 TheFieryLynx opened 3 years ago
4
update

#63 Topology1225 closed 3 years ago
1
fix REAME.md & specify a default path in tf2 example

#62 MariaMsu closed 3 years ago
1
How to use auto-attack with tensorflow?

#61 MariaMsu closed 3 years ago
2
update for pytorch 1.9.0, add zero_gradients function

#60 Topology1225 closed 3 years ago
2
ImportError: cannot import name 'zero_gradients' from 'torch.autograd.gradcheck'

#59 Topology1225 closed 3 years ago
0
Add [Stochastic LWTA]

#58 konpanousis opened 3 years ago
11