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
  • [41] Defending Deep Learning Based Anomaly Detection Systems Against White-Box Adversarial Examples and Backdoor Attacks
    Alrawashdeh, Khaled
    Goldsmith, Stephen
    PROCEEDINGS OF THE 2020 IEEE INTERNATIONAL SYMPOSIUM ON TECHNOLOGY AND SOCIETY (ISTAS), 2021, : 294 - 301
  • [42] Patch2Self: Denoising Diffusion MRI with Self-Supervised Learning
    Fadnavis, Shreyas
    Batson, Joshua
    Garyfallidis, Eleftherios
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020, 2020, 33
  • [43] Weakly Supervised Semantic Segmentation with Patch-Based Metric Learning Enhancement
    Chan, Patrick P. K.
    Chen, Keke
    Xu, Linyi
    Hu, Xiaoman
    Yeung, Daniel S.
    ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2021, PT III, 2021, 12893 : 471 - 482
  • [44] Patch-level Representation Learning for Self-supervised Vision Transformers
    Yun, Sukmin
    Lee, Hankook
    Kim, Jaehyung
    Shin, Jinwoo
    2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2022, : 8344 - 8353
  • [45] A Hierarchical Vision Transformer Using Overlapping Patch and Self-Supervised Learning
    Ma, Yaxin
    Li, Ming
    Chang, Jun
    2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN, 2023,
  • [46] Improvements to context based self-supervised learning
    Mundhenk, T. Nathan
    Ho, Daniel
    Chen, Barry Y.
    2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 9339 - 9348
  • [47] PatchZero: Defending against Adversarial Patch Attacks by Detecting and Zeroing the Patch
    Xu, Ke
    Xiao, Yao
    Zheng, Zhaoheng
    Cai, Kaijie
    Nevatia, Ram
    2023 IEEE/CVF WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV), 2023, : 4621 - 4630
  • [48] Backdoor attacks against distributed swarm learning
    Chen, Kongyang
    Zhang, Huaiyuan
    Feng, Xiangyu
    Zhang, Xiaoting
    Mi, Bing
    Jin, Zhiping
    ISA TRANSACTIONS, 2023, 141 : 59 - 72
  • [49] S2P-Matching: Self-Supervised Patch-Based Matching Using Transformer for Capsule Endoscopic Images Stitching
    Lu, Feng
    Zhou, Dao
    Chen, Haoyang
    Liu, Shuai
    Ling, Xianliang
    Zhu, Lei
    Gong, Tingting
    Sheng, Bin
    Liao, Xiaofei
    Jin, Hai
    Li, Ping
    Feng, David Dagan
    IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2025, 72 (02) : 540 - 551
  • [50] Gated Self-supervised Learning for Improving Supervised Learning
    Fuadi, Erland Hillman
    Ruslim, Aristo Renaldo
    Wardhana, Putu Wahyu Kusuma
    Yudistira, Novanto
    2024 IEEE CONFERENCE ON ARTIFICIAL INTELLIGENCE, CAI 2024, 2024, : 611 - 615