MadryLab/cifar10_challenge

Issues

• pretrained model link expired

  #30 opened 3 years ago by hank08tw
  1

• Submitting my result to white CIFAR-10 leaderboards

  #29 opened 4 years ago by liuye6666
  1

• lack of Sign function

  #28 opened 4 years ago by Sumching
  0

• The config of training robust model for CIFAR10.

  #27 opened 4 years ago by ylsung
  1

• pytorch definition of the model

  #26 opened 5 years ago by kartikgupta-at-anu
  1

• clean accuracy on adversarially trained cifar10 resnet 18

  #25 opened 5 years ago by wahrheit-git
  1

• Naturally trained network gives 78% test accuracy on pgd attack.

  #24 opened 5 years ago by baytasin
  1

• About the network architecture

  #23 opened 5 years ago by dongyp13
  1

• Making adversarial examples during training

  #22 opened 5 years ago by symoon11
  1

• Overflow when random_restart is false

  #21 opened 5 years ago by hanboa
  1

• Number of trainable parameters

  #20 opened 5 years ago by JonathanCMitchell
  1

• When generating uniform noise in random start, floating point number will cause invalid pixel value.

  #19 opened 5 years ago by Line290
  1

• Matching training statistics

  #16 opened 5 years ago by JonathanCMitchell
  2

• PGD steps along the sign of the gradient

  #18 opened 5 years ago by SohamTamba
  1

• About the convergence of training.

  #17 opened 5 years ago by anonymous530
  2

• Dataset normalization

  #15 opened 5 years ago by TimurIbrayev
  2

• Same accuracy logged twice

  #14 opened 5 years ago by JonathanCMitchell
  1

• Base network questions and implementation.

  #13 opened 5 years ago by JonathanCMitchell
  3

• Googlenet with owndata

  #12 opened 6 years ago by rajasekharponakala
  1

• Image Channels

  #11 opened 6 years ago by rajasekharponakala
  1

• About the accuracy of adversarial examples

  #10 opened 6 years ago by lith0613
  1

• about the loss in the pgd_attack

  #9 opened 6 years ago by lith0613
  1

• White-box result of madry_lab_challenges in examples of cleverhans.

  #8 opened 6 years ago by lepangdan
  1

• config for 20 iteration pgd that gives 47.04% accuracy.

  #7 opened 6 years ago by msingh27
  1

• Questions about recreating paper results

  #6 opened 6 years ago by inkawhich
  4

• How to determine "best" model

  #5 opened 6 years ago by inkawhich
  4

• Image out of valid range for the first iteration of PGD attack

  #4 opened 6 years ago by vipinpillai
  2

• cifar_input.py Function get_next_batch() has a small bug.

  #3 opened 6 years ago by Mrxiaoyuer
  1