LukasStruppek/Plug-and-Play-Attacks
[ICML 2022 / ICLR 2024] Source code for our papers "Plug & Play Attacks: Towards Robust and Flexible Model Inversion Attacks" and "Be Careful What You Smooth For".
This project helps privacy and security researchers evaluate how much sensitive information a machine learning model might be leaking about its training data. It takes an existing image classifier and a pre-trained image generator (like StyleGAN2) as input, then produces synthetic images that reveal characteristic features from the model's training data. This tool is designed for researchers focused on model privacy, enabling them to test the robustness of their models against advanced inversion attacks and understand the implications of different training techniques like label smoothing.
No commits in the last 6 months.
Use this if you are a machine learning researcher or privacy engineer who needs to assess the privacy leakage of image classification models by generating class-representative samples that mimic the characteristics of private training data.
Not ideal if you are looking for a general-purpose image generation tool or a method to anonymize datasets directly, as this tool is specifically for auditing model privacy via inversion attacks.
Stars
46
Forks
13
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jul 18, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/LukasStruppek/Plug-and-Play-Attacks"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
QData/TextAttack
TextAttack 🐙 is a Python framework for adversarial attacks, data augmentation, and model...
ebagdasa/backdoors101
Backdoors Framework for Deep Learning and Federated Learning. A light-weight tool to conduct...
THUYimingLi/backdoor-learning-resources
A list of backdoor learning resources
zhangzp9970/MIA
Unofficial pytorch implementation of paper: Model Inversion Attacks that Exploit Confidence...
VinAIResearch/Warping-based_Backdoor_Attack-release
WaNet - Imperceptible Warping-based Backdoor Attack (ICLR 2021)