Florian Tramer

Steal-ML. Model extraction attacks on Machine-Learning-as-a-Service platforms.

LM_Memorization. Training data extraction on GPT-2

slalom. Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware

ensemble-adv-training. Ensemble Adversarial Training on MNIST

Handcrafted-DP. Python

MultiRobustness. Code for the paper "Adversarial Training and Robustness for Multiple Perturbations", NeurIPS 2019

ad-versarial. Python

Excessive-Invariance. Source code for the paper "Exploiting Excessive Invariance caused by Norm-Bounded Adversarial Robustness"

FaceCure. Python

sealed-glass-proofs. C

stanford-suns-nips17. Code used by the team Stanford&Suns for the NIPS 2017 Competition on Adversarial Examples (https://www.kaggle.com/c/nips-2017-defense-against-adversarial-attack)