softsys4ai/athena

Issues

• Implementing some random noise as a weak attack

  #8 opened 3 years ago by pooyanjamshidi
  1

• Organize the defense in a folder

  #23 opened 3 years ago by MENG2010
  1

• Readme for the project showing the end top end workflow

  #19 opened 3 years ago by pooyanjamshidi
  1

• refactor

  #32 opened 3 years ago by MENG2010
  1

• reset function in src/evaluation/eval_whitebox.py

  #36 opened 4 years ago by ScottLiao920
  5

• End-to-end demo for our defense approach

  #25 opened 4 years ago by oceank
  1

• Normalized l2-dissimilarity

  #35 opened 5 years ago by ScottLiao920
  4

• Investigation on the difference of inference time between BS and AE

  #10 opened 5 years ago by oceank
  1

• blackbox attack

  #3 opened 5 years ago by MENG2010
  1

• CW attack

  #6 opened 5 years ago by MENG2010
  1

• Accuracy of some transform models are lower than that of clean model

  #14 opened 5 years ago by oceank
  1

• Visualizing label separateability using t-sne

  #20 opened 5 years ago by pooyanjamshidi
  1

• Evaluate White-box Threat Model

  #28 opened 5 years ago by MENG2010
  1

• Evaluate Grey-box Threat Model

  #29 opened 5 years ago by MENG2010
  1

• Support MIM attack

  #26 opened 5 years ago by MENG2010
  1

• Get scripts run on rci nodes.

  #2 opened 5 years ago by MENG2010
  4

• construct tests for this project

  #7 opened 5 years ago by MENG2010
  0

• pgd attack

  #4 opened 5 years ago by MENG2010
  1

• DEEPFOOL (l_inf norm)

  #5 opened 5 years ago by MENG2010
  0

• Issues running script on gcloud

  #9 opened 5 years ago by MENG2010
  3

• support one-pixel attack

  #1 opened 5 years ago by MENG2010
  1

• Investigate how adv attacks change the class activation mapping (CAM)

  #21 opened 5 years ago by pooyanjamshidi
  1

• Test if the current defenses does not reply on a specifc AE type

  #12 opened 5 years ago by oceank
  2

• make current defense approaches (ensemble models) to not depend on a specific AE type

  #13 opened 5 years ago by oceank
  1

• exceptions were thrown when applying some sort of transformations

  #17 opened 5 years ago by MENG2010
  1

• poisson-noise transformation model crashes when predicting the AE, jsma (theta-10, gamma-30)

  #18 opened 5 years ago by oceank
  2

• Tune and fix bugs for new transformations

  #22 opened 5 years ago by MENG2010
  7

• SC computing cluster: deploy a docker container that could access a GPU node

  #24 opened 5 years ago by oceank
  0

• Detection as a defense

  #11 opened 5 years ago by oceank
  1

• use FLAG to manage configurations

  #16 opened 5 years ago by MENG2010
  0

• manage the information with logging

  #15 opened 5 years ago by MENG2010
  0