yilinjuang / Adversarial-Learning

Individual Study under Prof. S. D. Lin
1 stars 0 forks source link

Adversarial-Learning

Individual Study under Prof. S. D. Lin.

Requirements

Black-box Attack

Forked from mnist_tutorial_tf. Use FGSM attack against ensemble methods including

Result

Raw Data Adaptive Boosting Accuracy-Max_Depth Adaptive Boosting Accuracy-N_Estimators Gradient Boosting Accuracy-Max_Depth Gradient Boosting Accuracy-N_Estimators Random Forest Accuracy-Max_Depth Random Forest Accuracy-N_Estimators