yuxiangw/autodp

Issues

• An issue when I installed "autodp": Preparing metadata (setup.py) ... error

  #48 opened a year ago by kangzhai
  2

• the noise is much greater than the probability. How should I use it correctly?

  #50 opened a year ago by szzhh
  0

• Does autodp support arbitrary group size?

  #45 opened a year ago by thupchnsky
  2

• Update Installation Instructions

  #42 opened 2 years ago by kiddyboots216
  3

• `bounds` cannot be used together with `method=Brent` in latest version of scipy (>= v1.10.1)

  #40 opened 2 years ago by thupchnsky
  4

• Composing different mechanism with different sensitivies

  #35 opened 2 years ago by SaitejaUtpala
  5

• AFA of composition of subsampled Laplace Mechanism breaks down

  #34 opened 2 years ago by zhuyizheng
  1

• Looseness in analytic Gaussian mechanism?

  #39 opened 2 years ago by ryan112358
  3

• Can't install in GBK locale

  #37 opened 2 years ago by ZisIsNotZis
  1

• Can PATE be used in knowledge distillation to calculate privacy budgets?

  #41 opened 2 years ago by w-yd
  0

• About the bisection method used for converting RDP to approximate DP

  #36 opened 2 years ago by spliew
  1

• Amplification with sampling without replacments is throwing following error.

  #33 opened 3 years ago by SaitejaUtpala
  4

• Issue with SSP_scale and AdaSSP_scale inheritance

  #22 opened 3 years ago by pvarshney1729
  2

• Slow privacy calibration

  #6 opened 3 years ago by ryan112358
  0

• Pure Fdp gaussian mechanism doesn't work under composition of multiple rounds

  #15 opened 4 years ago by AHabes
  1

• documentation question

  #11 opened 4 years ago by arakotom
  2

• issue with "privacy_calibrator.subsample_epsdelta_inverse(eps,delta,prob=gamma)"

  #2 opened 4 years ago by dadelani
  1

• huge fan of this work

  #7 opened 4 years ago by iamtrask
  0

• difference between eps using this method and abadi

  #3 opened 5 years ago by srxzr
  3