Defending Against Patch-based Backdoor Attacks on Self-Supervised Learning

被引:5
|
作者
Tejankar, Ajinkya [1 ]
Sanjabi, Maziar [2 ]
Wang, Qifan [2 ]
Wang, Sinong [2 ]
Firooz, Hamed [2 ]
Pirsiavash, Hamed [1 ]
Tan, Liang [2 ]
机构
[1] Univ Calif Davis, Davis, CA 95616 USA
[2] Meta AI, Delaware, OH USA
关键词
D O I
10.1109/CVPR52729.2023.01178
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Recently, self-supervised learning (SSL) was shown to be vulnerable to patch-based data poisoning backdoor attacks. It was shown that an adversary can poison a small part of the unlabeled data so that when a victim trains an SSL model on it, the final model will have a backdoor that the adversary can exploit. This work aims to defend self-supervised learning against such attacks. We use a three-step defense pipeline, where we first train a model on the poisoned data. In the second step, our proposed defense algorithm (PatchSearch) uses the trained model to search the training data for poisoned samples and removes them from the training set. In the third step, a final model is trained on the cleaned-up training set. Our results show that PatchSearch is an effective defense. As an example, it improves a model's accuracy on images containing the trigger from 38.2% to 63.7% which is very close to the clean model's accuracy, 64.6%. Moreover, we show that PatchSearch outperforms baselines and state-of-the-art defense approaches including those using additional clean, trusted data. Our code is available at https://github.com/UCDvision/PatchSearch
引用
收藏
页码:12239 / 12249
页数:11
相关论文
共 50 条
  • [21] Unsupervised anomaly detection for posteroanterior chest X-rays using multiresolution patch-based self-supervised learning
    Kim, Minki
    Moon, Ki-Ryum
    Lee, Byoung-Dai
    SCIENTIFIC REPORTS, 2023, 13 (01)
  • [22] Defending against Backdoor Attacks in Natural Language Generation
    Sun, Xiaofei
    Li, Xiaoya
    Meng, Yuxian
    Ao, Xiang
    Lyu, Lingjuan
    Li, Jiwei
    Zhang, Tianwei
    THIRTY-SEVENTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 37 NO 4, 2023, : 5257 - 5265
  • [23] Invariant Aggregator for Defending against Federated Backdoor Attacks
    Wang, Xiaoyang
    Dimitriadis, Dimitrios
    Koyejo, Sanmi
    Tople, Shruti
    INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 238, 2024, 238
  • [24] TRAINING SET CLEANSING OF BACKDOOR POISONING BY SELF-SUPERVISED REPRESENTATION LEARNING
    Wang, Hang
    Karami, Sahar
    Dia, Ousmane
    Ritter, Hippolyt
    Emamjomeh-Zadeh, Ehsan
    Chen, Jiahui
    Xiang, Zhen
    Miller, David J.
    Kesidis, George
    arXiv, 2022,
  • [25] Training Set Cleansing of Backdoor Poisoning by Self-Supervised Representation Learning
    Wang, Hang
    Karami, Sahar
    Dia, Ousmane
    Ritter, Hippolyt
    Emamjomeh-Zadeh, Ehsan
    Chen, Jiahui
    Xiang, Zhen
    Miller, David J.
    Kesidis, George
    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2023,
  • [26] Towards defending adaptive backdoor attacks in Federated Learning
    Yang, Han
    Gu, Dongbing
    He, Jianhua
    ICC 2023-IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS, 2023, : 5078 - 5084
  • [27] Effective Targeted Attacks for Adversarial Self-Supervised Learning
    Kim, Minseon
    Ha, Hyeonjeong
    Son, Sooel
    Hwang, Sung Ju
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [28] A Self Supervised Defending Mechanism Against Adversarial Iris Attacks based on Wavelet Transform
    Meenakshi, K.
    Maragatham, G.
    INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS, 2021, 12 (02) : 564 - 569
  • [29] Defending against Insertion-based Textual Backdoor Attacks via Attribution
    Li, Jiazhao
    Wu, Zhuofeng
    Ping, Wei
    Xiao, Chaowei
    Vydiswaran, V. G. Vinod
    FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2023), 2023, : 8818 - 8833
  • [30] FLSAD: Defending Backdoor Attacks in Federated Learning via Self-Attention Distillation
    Chen, Lucheng
    Liu, Xiaoshuang
    Wang, Ailing
    Zhai, Weiwei
    Cheng, Xiang
    SYMMETRY-BASEL, 2024, 16 (11):