/ml-leaks-pytorch

Implementation of ML Leaks paper on Pytorch

Primary LanguagePython

ml-leaks-pytorch

Implementation of ML Leaks paper on Pytorch

This is a pytorch implementation of the first adversary attack, which is explained on ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Models, here https://arxiv.org/abs/1806.01246

On the first attack the adversary uses one shadow model, that is constructed the same way as the target, to train the attack model.