Jianbo-Lab/HSJA

Issues

• Question about delta

  #18 opened a year ago
  1

• Why this line code exists? HELP plz

  #17 opened 2 years ago
  1

• Unnecessary binary search?

  #16 opened 3 years ago
  0

• Hi，I want to know which represents the number of queries in hsja.py？

  #15 opened 3 years ago
  3

• Which true gradient does HSJA estimated for? (in function "approximate_gradient")

  #14 opened 3 years ago
  0

• MNIST model for attack

  #13 opened 3 years ago
  0

• Provide parameters and example models for ImageNet

  #12 opened 3 years ago
  0

• a problem about convergence

  #11 opened 3 years ago
  0

• this is a problem to run this code ,but i don't how to solve , please query a help,"construct_original_network grads = tape.gradient(preds[:,c], x) File "C:\Users\高婷\AppData\Roaming\Python\Python37\site-packages\tensorflow\python\eager\backprop.py", line 1086, in gradient unconnected_gradients=unconnected_gradients) File "C:\Users\高婷\AppData\Roaming\Python\Python37\site-packages\tensorflow\python\eager\imperative_grad.py", line 77, in imperative_grad compat.as_str(unconnected_gradients.value)) AttributeError: 'KerasTensor' object has no attribute '_id'"

  #10 opened 4 years ago
  0

• What does baseline substraction mean in the gradient estimation?

  #9 opened 4 years ago
  0

• Question about L_infinity norm attack of HSJA.

  #8 opened 4 years ago
  1

• Why the perturbed image is not clip with min=0 and max=1 after binary search step? Bug?

  #7 opened 4 years ago
  1

• Which stepsize_search performs better in terms of query and distortion?

  #6 opened 4 years ago
  1

• How is input norm calculated?

  #5 opened 4 years ago
  2

• Is it possible and easy to enable batch processing?

  #4 opened 4 years ago
  2

• The approximate_gradient function seems not follow the equations in paper

  #3 opened 5 years ago
  10

• HSJA is extremely slow when I use the params in this repo

  #2 opened 5 years ago
  12

• The value of theta is different from the paper

  #1 opened 5 years ago
  2