EN 520.655 Project 1

Nisarg A. Shah¹ {[email protected]} and Yasiru Ranasinghe¹ {[email protected]}

¹ Johns Hopkins University

Prerequisites

• Python 3.6+
• PyTorch 1.0+

Training

# Start training with: 
python main.py

# You can manually resume the training with: 
python main.py --resume --lr=0.01

Fast Gradient Signed Method attack

# Start training with:
python fgsm.py

Projected Gradient Descent L-2 norm attack

# Start training with:
python pgd_l2.py

Projected Gradient Descent L-infinity norm attack

# Start training with:
python pgd_linf.py

Carlini-Wagner attack

# Start training with:
python cw.py

Adversarial training

# Start training with:
python adversarial_training.py