zhoumingyi/ModelObfuscator
Code for our paper "Modelobfuscator: Obfuscating Model Information to Protect Deployed ML-Based Systems" that has been published by ISSTA'23
This tool helps mobile app developers protect the machine learning models embedded directly within their apps on edge devices. It takes a deployed TFLite model as input and produces an obfuscated version of that model, making it much harder for attackers to extract sensitive information like its structure or training data. Mobile app developers concerned about intellectual property theft or adversarial attacks on their on-device AI can use this.
No commits in the last 6 months.
Use this if you are deploying a machine learning model directly within a mobile application or on an edge device and need to safeguard it against white-box attacks and reverse engineering.
Not ideal if your machine learning model is hosted on a secure cloud server and not directly exposed on an end-user device.
Stars
19
Forks
7
Language
C++
License
MIT
Category
Last pushed
May 18, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/zhoumingyi/ModelObfuscator"
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...
liuyugeng/ML-Doctor
Code for ML Doctor