https://github.com/duoergun0729/miaa

code of "Improving Membership Inference Attack Robustness With Attention Against Adversarial Examples"
https://github.com/duoergun0729/miaa

Last synced: about 2 months ago
JSON representation

code of "Improving Membership Inference Attack Robustness With Attention Against Adversarial Examples"

• Host: GitHub
• URL: https://github.com/duoergun0729/miaa
• Owner: duoergun0729
• License: apache-2.0
• Created: 2019-12-30T22:06:13.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2020-01-03T07:02:14.000Z (almost 5 years ago)
• Last Synced: 2024-10-31T16:03:38.902Z (2 months ago)
• Language: Jupyter Notebook
• Size: 767 KB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# MIAA

code of "Improving Membership Inference Attack Robustness With Attention Against Adversarial Examples"

Code modification based on "MemGuard: Defending against Black-Box Membership Inference Attacks via Adversarial Examples" [code](https://github.com/jjy1994/MemGuard)

Required python tool:

- tensorflow-gpu==1.8.0

- python==2.7

- pandas==0.24.2

- Keras==2.1.6

- Keras-Applications==1.0.4

- Keras-Preprocessing==1.0.2

# Dataset description

# Code usage

    python main_run.py

result saved at log/result.log

# Citation

If you use this code, please cite the following paper:

# Acknowledgment

We would like to thank Jinyuan Jia and Neil Zhenqiang Gong (Duke University) for their open source code of [MemGuard](https://github.com/jjy1994/MemGuard).