fra31/auto-attack

Issues

• N_classes in targeted attacks

  #110 opened a month ago by MaxHeuillet
  1

• How can I obtain your pre-trained weights?

  #109 opened 2 months ago by tianhhhh
  2

• Parallelized computing

  #94 opened 3 years ago by Bai-YT
  6

• DLR Loss and Zero Gradient

  #108 opened 4 months ago by SerezD
  1

• Unable to use Python debugger

  #88 opened 3 years ago by yuezuegu
  4

• About `n_iter`, to align with others

  #106 opened 10 months ago by userElaina
  4

• normalization with deepfool fmodel

  #103 opened a year ago by jS5t3r
  4

• max Linf perturbation is larger than the epsilon

  #86 opened 2 years ago by XiaoYaoYouUSTC
  3

• Softmax probabilities instead of logits

  #102 opened 2 years ago by fugokidi
  3

• Can AutoAttack be used to dense prediction task?

  #100 opened 2 years ago by zhanglijun95
  2

• TF1: How do labels get modified by batch_datapoint_idcs?

  #99 opened 2 years ago by giorgiopiras
  8

• Using the tool for general interval of input region

  #97 opened 2 years ago by shubhamugare
  3

• Invalid configuration for square attack

  #95 opened 3 years ago by CNOCycle
  2

• Unable to use in TF2

  #89 opened 3 years ago by Shubhajit1988
  5

• gradient computation issue

  #87 opened 3 years ago by CNOCycle
  4

• ensemble attack

  #85 opened 3 years ago by Jialiang14
  3

• Autoattack and APGDT

  #52 opened 4 years ago by Yeez-lee
  3

• "The connection to the C10d store has failed" on distributed evaluation

  #84 opened 3 years ago by chawins
  2

• Problem that forward output is tuple rather than tensor

  #83 opened 3 years ago by momo1986
  1

• A Bug in Square Attack

  #81 opened 3 years ago by GuanlinLee
  4

• 4 classes model; targeted attacks are different

  #79 opened 3 years ago by jS5t3r
  2

• Import error for pytorch version 1.10.0

  #78 opened 3 years ago by henrikju
  3

• eps 8./255. works fine 4./255. does not work fine.

  #76 opened 3 years ago by jS5t3r
  12

• Update AutoAttack Framework to pytorch 1.9

  #68 opened 3 years ago by jS5t3r
  2

• Question about variation in reported (clean & robust accuracy) metrics

  #74 opened 3 years ago by choltz95
  2

• Regarding the checks recently added to AutoAttack

  #73 opened 4 years ago by zangobot
  4

• The DLR loss is undefined for classification problems with fewer than 4 classes

  #70 opened 4 years ago by peck94
  1

• Add FSGM_APR_SP

  #69 opened 4 years ago by iCGY96
  4

• Add LBGAT on CIFAR-10

  #36 opened 4 years ago by jiequancui
  3

• Add LBGAT

  #34 opened 4 years ago by jiequancui
  8

• Adversarials are equals to originals

  #64 opened 4 years ago by TheFieryLynx
  4

• ImportError: cannot import name 'zero_gradients' from 'torch.autograd.gradcheck'

  #59 opened 4 years ago by Topology1225
  0

• How to use auto-attack with tensorflow?

  #61 opened 4 years ago by MariaMsu
  2

• Default configuration of CIFAR100

  #57 opened 4 years ago by CNOCycle
  4

• Add [Stochastic LWTA]

  #58 opened 4 years ago by konpanousis
  11

• Modified Square Attack to Randomized Defense

  #56 opened 4 years ago by sungyoon-lee
  1

• Why do you evaluate your model without standard mean and standard variance normalization?

  #55 opened 4 years ago by jS5t3r
  1

• Support FP16 in pytorch

  #51 opened 4 years ago by CNOCycle
  4

• Random seed

  #54 opened 4 years ago by kaustubhsridhar
  4

• Normalization of CIFAR-10 images

  #53 opened 4 years ago by GeJulia
  2

• Evaluation of Defended Models

  #50 opened 4 years ago by akshayag
  1

• Normalized input data for AutoAttack

  #46 opened 4 years ago by Liang-Shang
  2

• What should be reported, the average of results or the minimum?

  #45 opened 4 years ago by Hrushikesh-github
  1

• Rounding to nearest pixel value breaks almost all attacks

  #44 opened 4 years ago by huyvnphan
  3

• Guidelines for evaluating defense using Auto Attack

  #43 opened 4 years ago by SohamTamba
  2

• Add Entropic Retraining

  #41 opened 4 years ago by spp01
  4

• Add Entropic Retraining

  #42 opened 4 years ago by spp01
  0

• Width-Adjusted-Regularization Update

  #40 opened 4 years ago by tabrisweapon
  2

• DLR loss implementation in TF2

  #37 opened 4 years ago by CNOCycle
  2

• A missing bracket in code

  #33 opened 4 years ago by THULimy
  0