Papers

Updating...

Attack

First adversarial example: https://arxiv.org/pdf/1312.6199.pdf

Futher explanation: https://arxiv.org/pdf/1412.6572.pdf

Transferability: https://arxiv.org/pdf/1605.07277.pdf

blackbox-attack: https://arxiv.org/pdf/1609.02943.pdf

Poisoning: https://arxiv.org/pdf/1804.00308.pdf

Model-stealing: https://arxiv.org/pdf/1804.00308.pdf

Defense

Input restruction: https://arxiv.org/pdf/1412.5068.pdf https://arxiv.org/pdf/1711.00117.pdf

DNN verification: https://arxiv.org/abs/1811.07108

Network Distillation: https://arxiv.org/pdf/1511.04508.pdf

Differential privacy: https://arxiv.org/pdf/1607.00133.pdf

Training data filter: https://arxiv.org/pdf/1606.01584.pdf

Input preprocessing: https://arxiv.org/pdf/1710.00942.pdf

PATE: https://arxiv.org/pdf/1610.05755.pdf

source code:

Paper with code: https://paperswithcode.com/

Pytorch start-up: https://pytorch123.com/

Cleverhans for adversarial example: https://github.com/cleverhans-lab/cleverhans

Code adversarial example paper: https://arxiv.org/pdf/1711.00117.pdf