MinChen00/UnlearningLeaks
Official implementation of "When Machine Unlearning Jeopardizes Privacy" (ACM CCS 2021)
This project helps evaluate the privacy risks associated with machine unlearning techniques, especially concerning membership inference attacks. It takes in various datasets and trained machine learning models, then performs attacks to determine if specific training data points can be identified, even after "unlearning." Data privacy researchers, machine learning engineers, and security analysts can use this to assess the effectiveness and security implications of different unlearning strategies.
No commits in the last 6 months.
Use this if you are a researcher or practitioner in machine learning and data privacy, looking to test how secure your unlearning methods are against sophisticated privacy attacks.
Not ideal if you are looking for a tool to implement machine unlearning in your production systems or to protect data from general security threats.
Stars
50
Forks
6
Language
Python
License
GPL-3.0
Category
Last pushed
May 20, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/MinChen00/UnlearningLeaks"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
google/scaaml
SCAAML: Side Channel Attacks Assisted with Machine Learning
pralab/secml
A Python library for Secure and Explainable Machine Learning
Koukyosyumei/AIJack
Security and Privacy Risk Simulator for Machine Learning (arXiv:2312.17667)
AI-SDC/SACRO-ML
Collection of tools and resources for managing the statistical disclosure control of trained...
oss-slu/mithridatium
Mithridatium is a research-driven project aimed at detecting backdoors and data poisoning in...