Backdoor Learning Resources

This Github repository summarizes a list of Backdoor Learning resources. For more details and the categorization criteria, please refer to our survey.

We will try our best to continuously maintain this Github Repository in a monthly manner.

Why Backdoor Learning?

Backdoor learning is an emerging research area, which discusses the security issues of the training process towards machine learning algorithms. It is critical for safely adopting third-party training resources or models in reality.

Note: 'Backdoor' is also commonly called the 'Neural Trojan' or 'Trojan'.

News

• 2023/7/24: I add ten ICLR'23 papers. All papers from this conference should have been included now.
• 2023/7/23: I add seven NeurIPS'22 papers and four AAAI'23 papers. All papers from these conferences should have been included now.
• 2023/01/25: I am deeply sorry that I have recently suspended the reading of related papers and the updates of this Repo, due to some personal issues such as sickness and writing Ph.D. dissertation. I will restart the update after June 2023.
• 2022/12/05: I slightly change the repo format by placing conference papers before journal papers. Specifically, in the same year, please place the conference paper before the journal paper, as journals are usually submitted a long time ago and therefore have some lag.
• 2022/12/05: I add three ECCV'22 papers. All papers from this conference should have been included now.

Reference

If our repo or survey is useful for your research, please cite our paper as follows:

@article{li2022backdoor,
  title={Backdoor learning: A survey},
  author={Li, Yiming and Jiang, Yong and Li, Zhifeng and Xia, Shu-Tao},
  journal={IEEE Transactions on Neural Networks and Learning Systems},
  year={2022}
}

Contributing

<p align="center"> <img src="http://cdn1.sportngin.com/attachments/news_article/7269/5172/needyou_small.jpg" alt="We Need You!"> </p>

Please help to contribute this list by contacting me or add pull request

Markdown format:

- Paper Name. 
  [[pdf]](link) 
  [[code]](link)
  - Author 1, Author 2, **and** Author 3. *Conference/Journal*, Year.

Note: In the same year, please place the conference paper before the journal paper, as journals are usually submitted a long time ago and therefore have some lag. (i.e., Conferences-->Journals-->Preprints)

Table of Contents

• Survey
• Toolbox
• Dissertation and Thesis
• Image and Video Classification
  • Poisoning-based Attack
  • Non-poisoning-based Attack
    • Weights-oriented Attack
    • Structure-modified Attack
    • Other Attacks
  • Backdoor Defense
    • Preprocessing-based Empirical Defense
    • Model Reconstruction based Empirical Defense
    • Trigger Synthesis based Empirical Defense
    • Model Diagnosis based Empirical Defense
    • Poison Suppression based Empirical Defense
    • Sample Filtering based Empirical Defense
    • Certificated Defense
• Attack and Defense Towards Other Paradigms and Tasks
  • Federated Learning
  • Transfer Learning
  • Reinforcement Learning
  • Semi-Supervised and Self-Supervised Learning
  • Quantization
  • Natural Language Processing
  • Graph Neural Networks
  • Point Cloud
  • Acoustics Signal Processing
  • Medical Science
  • Detection and Tracking
  • Vision Transformer
  • Diffusion Model
  • Cybersecurity
  • Others
• Evaluation and Discussion
• Backdoor Attack for Positive Purposes
• Competition

Survey

• Backdoor Learning: A Survey. [pdf]

  • Yiming Li, Yong Jiang, Zhifeng Li, and Shu-Tao Xia. IEEE Transactions on Neural Networks and Learning Systems, 2022.

• Backdoor Attacks and Countermeasures on Deep Learning: A Comprehensive Review. [pdf]

  • Yansong Gao, Bao Gia Doan, Zhi Zhang, Siqi Ma, Anmin Fu, Surya Nepal, and Hyoungshick Kim. arXiv, 2020.

• Data Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defenses. [pdf]

  • Micah Goldblum, Dimitris Tsipras, Chulin Xie, Xinyun Chen, Avi Schwarzschild, Dawn Song, Aleksander Madry, Bo Li, and Tom Goldstein. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022.

• A Comprehensive Survey on Poisoning Attacks and Countermeasures in Machine Learning. [link]

  • Zhiyi Tian, Lei Cui, Jie Liang, and Shui Yu. ACM Computing Surveys, 2022.

• Backdoor Attacks and Defenses in Federated Learning: State-of-the-art, Taxonomy, and Future Directions. [link]

  • Xueluan Gong, Yanjiao Chen, Qian Wang, and Weihan Kong. IEEE Wireless Communications, 2022.

• Backdoor Attacks on Image Classification Models in Deep Neural Networks. [link]

  • Quanxin Zhang, Wencong Ma, Yajie Wang, Yaoyuan Zhang, Zhiwei Shi, and Yuanzhang Li. Chinese Journal of Electronics, 2022.

• Defense against Neural Trojan Attacks: A Survey. [link]

  • Sara Kaviani and Insoo Sohn. Neurocomputing, 2021.

• A Survey on Neural Trojans. [pdf]

  • Yuntao Liu, Ankit Mondal, Abhishek Chakraborty, Michael Zuzak, Nina Jacobsen, Daniel Xing, and Ankur Srivastava. ISQED, 2020.

• Backdoor Attacks against Voice Recognition Systems: A Survey. [pdf]

  • Baochen Yan, Jiahe Lan, and Zheng Yan. arXiv, 2023.

• A Survey of Neural Trojan Attacks and Defenses in Deep Learning. [pdf]

  • Jie Wang, Ghulam Mubashar Hassan, and Naveed Akhtar. arXiv, 2022.

• Threats to Pre-trained Language Models: Survey and Taxonomy. [pdf]

  • Shangwei Guo, Chunlong Xie, Jiwei Li, Lingjuan Lyu, and Tianwei Zhang. arXiv, 2022.

• An Overview of Backdoor Attacks Against Deep Neural Networks and Possible Defences. [pdf]

  • Wei Guo, Benedetta Tondi, and Mauro Barni. arXiv, 2021.

• Deep Learning Backdoors. [pdf]

  • Shaofeng Li, Shiqing Ma, Minhui Xue, and Benjamin Zi Hao Zhao. arXiv, 2020.

Toolbox

• BackdoorBox
• TrojanZoo
• OpenBackdoor
• Backdoor Toolbox
• BackdoorBench
• backdoors101
• ART

Dissertation and Thesis

• Poisoning-based Backdoor Attacks in Computer Vision. [pdf]

  • Yiming Li. Ph.D. Dissertation at Tsinghua University, 2023.

• Defense of Backdoor Attacks against Deep Neural Network Classifiers. [pdf]

  • Zhen Xiang. Ph.D. Dissertation at The Pennsylvania State University, 2022.

• Towards Adversarial and Backdoor Robustness of Deep Learning. [link]

  • Yifan Guo. Ph.D. Dissertation at Case Western Reserve University, 2022.

• Toward Robust and Communication Efficient Distributed Machine Learning. [pdf]

  • Hongyi Wang. Ph.D. Dissertation at University of Wisconsin–Madison, 2021.

• Towards Robust Image Classification with Deep Learning and Real-Time DNN Inference on Mobile. [pdf]

  • Pu Zhao. Ph.D. Dissertation at Northeastern University, 2021.

• Countermeasures Against Backdoor, Data Poisoning, and Adversarial Attacks. [pdf]

  • Henry Daniel. Ph.D. Dissertation at University of Texas at San Antonio, 2021.

• Understanding and Mitigating the Impact of Backdooring Attacks on Deep Neural Networks. [pdf]

  • Kang Liu. Ph.D. Dissertation at New York University, 2021.

• Un-fair trojan: Targeted Backdoor Attacks against Model Fairness. [pdf]

  • Nicholas Furth. Master Thesis at New Jersey Institute of Technology, 2022.

• Check Your Other Door: Creating Backdoor Attacks in the Frequency Domain. [[pdf]](file:///C:/Users/R