Awesome

Label-Consistent Backdoor Attacks

This repository contains the minimal PyTorch implementation of Label-Consistent Backdoor Attacks on CIFAR-10 dataset. The official Tensorflow implementation is here.

Requirements

• python 3.7
• pytorch 1.6.0
• numpy
• tabulate
• pyyaml
• tqdm

Usage

Step1: Train an Adversarially Robust Model

For fast adversarial training, please refer to fast_adversarial. I also provide a PGD adversarially pretrained ResNet-18 model in here. I get the pretrained model by running train_pgd.py. The parameters of PGD advesarial traning are the same with the paper, which in turn was adapted from cifar10_challenge.

Step2: Generate Adversarially Peturbed Dataset

python craft_adv_dataset.py --config config/craft.yaml --gpu 0

This will generate adversarially peturbed dataset to data/adv_dataset/craft.npz by PGD attack. The parameters of PGD attack are the same with the paper.

Step3: Train a Backdoored Model

python poison_train.py --config config/poison_train.yaml --gpu 0

Note: For simplicity, I use a randomly generated trigger instead of a less visible and four-corner trigger in the paper section 4.4 (Improving backdoor trigger design). And the parameters of poison training in config/poison_train.yaml are adapted from pytorch-cifar, which may be different from the paper. Please refer to the experimental setup in the paper Appendix A.