Official implementation of the paper "Backdoor Attacks on Self-Supervised Learning".

Overview

SSL-Backdoor

Abstract

Large-scale unlabeled data has allowed recent progress in self-supervised learning methods that learn rich visual representations. State-of-the-art self-supervised methods for learning representations from images (MoCo and BYOL) use an inductive bias that different augmentations (e.g. random crops) of an image should produce similar embeddings. We show that such methods are vulnerable to backdoor attacks where an attacker poisons a part of the unlabeled data by adding a small trigger (known to the attacker) to the images. The model performance is good on clean test images but the attacker can manipulate the decision of the model by showing the trigger at test time. Backdoor attacks have been studied extensively in supervised learning and to the best of our knowledge, we are the first to study them for self-supervised learning. Backdoor attacks are more practical in self-supervised learning since the unlabeled data is large and as a result, an inspection of the data to avoid the presence of poisoned data is prohibitive. We show that in our targeted attack, the attacker can produce many false positives for the target category by using the trigger at test time. We also develop a knowledge distillation based defense algorithm that succeeds in neutralizing the attack. Our code is available here: https://github.com/UMBCvision/SSL-Backdoor.

Paper

Backdoor Attacks on Self-Supervised Learning

Updates

  • 04/07/2021 - Poison generation code added.
  • 04/08/2021 - MoCo v2, BYOL code added.
  • 04/14/2021 - Jigsaw, RotNet code added.

Requirements

All experiments were run using the following dependencies.

  • python=3.7
  • pytorch=1.6.0
  • torchvision=0.7.0
  • wandb=0.10.21 (for BYOL)
  • torchnet=0.0.4 (for RotNet)

Optional

  • faiss=1.6.3 (for k-NN evaluation)

Create ImageNet-100 dataset

The ImageNet-100 dataset (random 100-class subset of ImageNet), commonly used in self-supervision benchmarks, was introduced in [1].

To create ImageNet-100 from ImageNet, use the provided script.

cd scripts
python create_imagenet_subset.py --subset imagenet100_classes.txt --full_imagenet_path <path> --subset_imagenet_path <path>

Poison Generation

To generate poisoned ImageNet-100 images, create your own configuration file. Some examples, which we use for our targeted attack experiments, are in the cfg directory.

  • You can choose the poisoning to be Targeted (poison only one category) or Untargeted
  • The trigger can be text or an image (We used triggers introduced in [2]).
  • The parameters of the trigger (e.g. location, size, alpha etc.) can be modified according to the experiment.
  • The poison injection rate for the training set can be modified.
  • You can choose which split to generate. "train" generates poisoned training data, "val_poisoned" poisons all the validation images for evaluation purpose. Note: The poisoned validation images are all resized and cropped to 224x224 before trigger pasting so that all poisoned images have uniform trigger size.
cd poison-generation
python generate_poison.py <configuration-file>

SSL Methods

Pytorch Custom Dataset

All images are loaded from filelists of the form given below.

<dir-name-1>/xxx.ext <target-class-index>
<dir-name-1>/xxy.ext <target-class-index>
<dir-name-1>/xxz.ext <target-class-index>

<dir-name-2>/123.ext <target-class-index>
<dir-name-2>/nsdf3.ext <target-class-index>
<dir-name-2>/asd932_.ext <target-class-index>

Evaluation

All evaluation scripts return confusion matrices for clean validation data and a csv file enumerating the TP and FP for each category.

MoCo v2 [3]

The implementation for MoCo is from https://github.com/SsnL/moco_align_uniform modified slightly to suit our experimental setup.

To train a ResNet-18 MoCo v2 model on ImageNet-100 on 2 NVIDIA GEFORCE RTX 2080 Ti GPUs:

cd moco
CUDA_VISIBLE_DEVICES=0,1 python main_moco.py \
                        -a resnet18 \
                        --lr 0.06 --batch-size 256 --multiprocessing-distributed \
                        --world-size 1 --rank 0 --aug-plus --mlp --cos --moco-align-w 0 \
                        --moco-unif-w 0 --moco-contr-w 1 --moco-contr-tau 0.2 \
                        --dist-url tcp://localhost:10005 \ 
                        --save-folder-root <path> \
                        --experiment-id <ID> <train-txt-file>

To train linear classifier on frozen MoCo v2 embeddings on ImageNet-100:

CUDA_VISIBLE_DEVICES=0 python eval_linear.py \
                        --arch moco_resnet18 \
                        --weights <SSL-model-checkpoint-path>\
                        --train_file <path> \
                        --val_file <path>

We use the linear classifier normalization from CompRess: Self-Supervised Learning by Compressing Representations which says "To reduce the computational overhead of tuning the hyperparameters per experiment, we standardize the Linear evaluation as following. We first normalize the features by L2 norm, then shift and scale each dimension to have zero mean and unit variance."

To evaluate linear classifier on clean and poisoned validation set: (This script loads the cached mean and variance from previous step.)

CUDA_VISIBLE_DEVICES=0 python eval_linear.py \
                        --arch moco_resnet18 \
                        --weights <SSL-model-checkpoint-path> \
                        --val_file <path> \
                        --val_poisoned_file <path> \
                        --resume <linear-classifier-checkpoint> \
                        --evaluate --eval_data <evaluation-ID> \
                        --load_cache

To run k-NN evaluation of frozen MoCo v2 embeddings on ImageNet-100 (faiss library needed):

CUDA_VISIBLE_DEVICES=0 python eval_knn.py \
                        -a moco_resnet18 \
                        --weights <SSL-model-checkpoint-path> \
                        --train_file <path> \
                        --val_file <path> \
                        --val_poisoned_file <path> \
                        --eval_data <evaluation-ID>

BYOL [4]

The implementation for BYOL is from https://github.com/htdt/self-supervised modified slightly to suit our experimental setup.

To train a ResNet-18 BYOL model on ImageNet-100 on 4 NVIDIA GEFORCE RTX 2080 Ti GPUs: (This scripts monitors the k-NN accuracy on clean ImageNet-100 dataset at regular intervals.)

cd byol
CUDA_VISIBLE_DEVICES=0,1,2,3 python -m train \
                                    --exp_id <ID> \
                                    --dataset imagenet --lr 2e-3 --emb 128 --method byol \
                                    --arch resnet18 --epoch 200 \
                                    --train_file_path <path> \
                                    --train_clean_file_path <path> 
                                    --val_file_path <path>
                                    --save_folder_root <path>

To train linear classifier on frozen BYOL embeddings on ImageNet-100:

CUDA_VISIBLE_DEVICES=0 python -m test --dataset imagenet \
                            --train_clean_file_path <path> \
                            --val_file_path <path> \
                            --emb 128 --method byol --arch resnet18 \
                            --fname <SSL-model-checkpoint-path>

To evaluate linear classifier on clean and poisoned validation set:

CUDA_VISIBLE_DEVICES=0 python -m test --dataset imagenet \
                            --val_file_path <path> \
                            --val_poisoned_file_path <path> \
                            --emb 128 --method byol --arch resnet18 \
                            --fname <SSL-model-checkpoint-path> \
                            --clf_chkpt <linear-classifier-checkpoint-path> \
                            --eval_data <evaluation-ID> --evaluate

Jigsaw [5]

The implementation for Jigsaw is our own Pytorch reimplementation based on the authors’ Caffe code https://github.com/MehdiNoroozi/JigsawPuzzleSolver modified slightly to suit our experimental setup. There might be some legacy Pytorch code, but that doesn't affect the correctness of training or evaluation. If you are looking for a recent Pytorch implementation of Jigsaw, https://github.com/facebookresearch/vissl is a good place to start.

To train a ResNet-18 Jigsaw model on ImageNet-100 on 1 NVIDIA GEFORCE RTX 2080 Ti GPU: (The code doesn't support Pytorch distributed training.)

cd jigsaw
CUDA_VISIBLE_DEVICES=0 python train_jigsaw.py \
                                --train_file <path> \
                                --val_file <path> \
                                --save <path>

To train linear classifier on frozen Jigsaw embeddings on ImageNet-100:

CUDA_VISIBLE_DEVICES=0 python eval_conv_linear.py \
                        -a resnet18 --train_file <path> \
                        --val_file <path> \
                        --save <path> \
                        --weights <SSL-model-checkpoint-path>

To evaluate linear classifier on clean and poisoned validation set:

CUDA_VISIBLE_DEVICES=0 python eval_conv_linear.py -a resnet18 \
                            --val_file <path> \
                            --val_poisoned_file <path> \
                            --weights <SSL-model-checkpoint-path> \
                            --resume <linear-classifier-checkpoint-path> \
                            --evaluate --eval_data <evaluation-ID>

RotNet [6]

The implementation for RotNet is from https://github.com/gidariss/FeatureLearningRotNet modified slightly to suit our experimental setup. There might be some legacy Pytorch code, but that doesn't affect the correctness of training or evaluation. If you are looking for a recent Pytorch implementation of RotNet, https://github.com/facebookresearch/vissl is a good place to start.

To train a ResNet-18 Jigsaw model on ImageNet-100 on 1 NVIDIA TITAN RTX GPU: (The code doesn't support Pytorch distributed training. Choose the experiment ID config file as required.)

cd rotnet
CUDA_VISIBLE_DEVICES=0 python main.py --exp <ImageNet100_RotNet_*> --save_folder <path>

To train linear classifier on frozen RotNet embeddings on ImageNet-100:

CUDA_VISIBLE_DEVICES=0 python main.py --exp <ImageNet100_LinearClassifiers_*> --save_folder <path>

To evaluate linear classifier on clean and poisoned validation set:

CUDA_VISIBLE_DEVICES=0 python main.py --exp <ImageNet100_LinearClassifiers_*> \
                            --save_folder <path> \
                            --evaluate --checkpoint=<epoch_num> --eval_data <evaluation-ID>

Acknowledgement

This material is based upon work partially supported by the United States Air Force under Contract No. FA8750‐19‐C‐0098, funding from SAP SE, NSF grant 1845216, and also financial assistance award number 60NANB18D279 from U.S. Department of Commerce, National Institute of Standards and Technology. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the United States Air Force, DARPA, or other funding agencies.

References

[1] Yonglong Tian, Dilip Krishnan, and Phillip Isola. Contrastive multiview coding. arXiv preprint arXiv:1906.05849,2019.

[2] Aniruddha Saha, Akshayvarun Subramanya, and Hamed Pirsiavash. Hidden trigger backdoor attacks. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 11957–11965, 2020.

[3] Chen, Xinlei, et al. "Improved baselines with momentum contrastive learning." arXiv preprint arXiv:2003.04297 (2020).

[4] Jean-Bastien Grill, Florian Strub, Florent Altch́e, and et al. Bootstrap your own latent - a new approach to self-supervised learning. In Advances in Neural Information Processing Systems, volume 33, pages 21271–21284, 2020.

[5] Noroozi, Mehdi, and Paolo Favaro. "Unsupervised learning of visual representations by solving jigsaw puzzles." European conference on computer vision. Springer, Cham, 2016.

[6] Spyros Gidaris, Praveer Singh, and Nikos Komodakis. Unsupervised representation learning by predicting image rotations. In International Conference on Learning Representations, 2018.

Citation

If you find our paper, code or models useful, please cite us using

@article{saha2021backdoor,
  title={Backdoor Attacks on Self-Supervised Learning},
  author={Saha, Aniruddha and Tejankar, Ajinkya and Koohpayegani, Soroush Abbasi and Pirsiavash, Hamed},
  journal={arXiv preprint arXiv:2105.10123},
  year={2021}
}

Questions/Issues

Please create an issue on the Github Repo directly or contact [email protected] for any questions about the code.

Owner
UMBC Vision
The Computer Vision Lab at the University of Maryland, Baltimore County (UMBC)
UMBC Vision
Separate handling of protected media in Django, with X-Sendfile support

Django Protected Media Django Protected Media is a Django app that manages media that are considered sensitive in a protected fashion. Not only does t

Cobus Carstens 46 Nov 12, 2022
EyeJo是一款自动化资产风险评估平台,可以协助甲方安全人员或乙方安全人员对授权的资产中进行排查,快速发现存在的薄弱点和攻击面。

EyeJo EyeJo是一款自动化资产风险评估平台,可以协助甲方安全人员或乙方安全人员对授权的资产中进行排查,快速发现存在的薄弱点和攻击面。 免责声明 本平台集成了大量的互联网公开工具,主要是方便安全人员整理、排查资产、安全测试等,切勿用于非法用途。使用者存在危害网络安全等任何非法行为,后果自负,作

429 Dec 31, 2022
HatSploit collection of generic payloads designed to provide a wide range of attacks without having to spend time writing new ones.

HatSploit collection of generic payloads designed to provide a wide range of attacks without having to spend time writing new ones.

EntySec 5 May 10, 2022
Omega - From Wordpress admin to pty

The Linux tool to automate the process of getting a pty once you got admin credentials in a Wordpress site. Keep in mind that right now Omega only can attack Linux hosts.

Ángel Heredia 12 Nov 09, 2022
BoobSnail allows generating Excel 4.0 XLM macro. Its purpose is to support the RedTeam and BlueTeam in XLM macro generation.

Follow us on Twitter! BoobSnail BoobSnail allows generating XLM (Excel 4.0) macro. Its purpose is to support the RedTeam and BlueTeam in XLM macro gen

STM Cyber 232 Nov 21, 2022
Scanning for CVE-2021-44228

Filesystem log4j_scanner for windows and Unix. Scanning for CVE-2021-44228, CVE-2021-45046, CVE-2019-17571 Requires a minimum of Python 2.7. Can be ex

Brett England 4 Jan 09, 2022
Simples brute forcer de diretorios para web pentest.

🦑 dirbruter Simples brute forcer de diretorios para web pentest. ❕ Atenção Não ataque sites privados. Isto é illegal. 🖥️ Pré-requisitos Ultima versã

Dio brando 6 Jan 22, 2022
Operational information regarding the vulnerability in the Log4j logging library.

Log4j Vulnerability (CVE-2021-44228) This repo contains operational information regarding the vulnerability in the Log4j logging library (CVE-2021-442

Nationaal Cyber Security Centrum (NCSC-NL) 1.9k Dec 26, 2022
Facebook Fast Cracking Tool With Python

Pro-Crack Facebook Fast Cracking Tool This is a multi-password‌ cracking tool that can help you hack facebook accounts very quickly Installation On Te

ReD H4CkeR 5 Feb 19, 2022
Cobalt Strike < 4.4 dos CVE-2021-36798

CVE-2021-36798 CVE-2021-36798 Cobalt Strike 4.3 dos 用法 python3 CVE-2021-36798.py BeaconURL 打瘫Cobalt Strike 只需要一个包 已测试 4.3 4.2 参考: https://labs.sent

37 Nov 09, 2022
Buff A simple BOF library I wrote under an hour to help me automate with BOF attack

What is Buff? A simple BOF library I wrote under an hour to help me automate with BOF attack. It comes with fuzzer and a generic method to generate ex

0x00 3 Nov 21, 2022
Security system to prevent Shoulder Surfing Attacks

Surf_Sec Security system to prevent Shoulder Surfing Attacks. REQUIREMENTS: Python 3.6+ XAMPP INSTALLED METHOD TO CONFIGURE PROJECT: Clone the repo to

Aman Anand 1 Jan 27, 2022
This project is all about building an amazing application that will help users manage their passwords and even generate new passwords for them

An amazing application that will help us manage our passwords and even generate new passwords for us.

1 Jan 23, 2022
A web-app helping to create strong passwords that are easy to remember.

This is a simple Web-App that demonstrates a method of creating strong passwords that are still easy to remember. It also provides time estimates how long it would take an attacker to crack a passwor

2 Jun 04, 2021
python写的一款免杀工具(shellcode加载器)BypassAV,国内杀软全过(windows denfend)

python写的一款免杀工具(shellcode加载器)BypassAV,国内杀软全过(windows denfend)

1frame 266 Jan 02, 2023
Collection Of Discord Hacking Tools / Fun Stuff / Exploits That Is Completely Made Using Python.

Venom Collection Of Discord Hacking Tools / Fun Stuff / Exploits That Is Completely Made Using Python. Report Bug · Request Feature Contributing Well,

PndaBoi 25 Dec 06, 2022
A collection of write-ups and solutions for Cyber FastTrack Spring 2021.

IMPORTANT: Please contact us before you use any styling or content shown here! Cyber FastTrack Spring 2021 / National Cyber Scholarship Competition -

Alice 48 Aug 28, 2022
Obfuscate your python code into a string of integers. De-obfuscate also supported.

int-obfuscator Obfuscate your python code into a string of integers. De-obfuscate also supported. How it works: Each printable character gets replaced

6 Nov 13, 2022