search

Trusted-AI / adversarial-robustness-toolbox

Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams
https://adversarial-robustness-toolbox.readthedocs.io/en/latest/
MIT License
4.7k stars 1.14k forks source link

Support for additional attack model types in inference attacks #2153

Closed abigailgold closed 11 months ago

abigailgold commented 1 year ago

Blackbox membership inference attack currently supports 3 model types: nn, rf and gb. Blackbox attribute inference attack currently supports 2 model types: nn and rf. Both could benefit from supporting additional attack model types (e.g., KNN, LR, etc.)

abigailgold commented 1 year ago

Also need to replace sklearn MLP implementation in attribute inference blackbox to pytorch (to improve performance).