Clean-label attack based on negative afterimage on neural networks

被引:0
作者
Zang, Liguang [1 ]
Li, Yuancheng [1 ]
机构
[1] North China Elect Power Univ, Dept Control & Comp Engineeringg, Beijing 102206, Peoples R China
关键词
Clean-label attack; Adversarial attacks; Afterimage; Machine learning;
D O I
10.1007/s13042-024-02230-3
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Adversarial attacks can fool machine learning models by adding small but carefully designed perturbations to an image, resulting in the model to misclassify it into another class. In general, an attacker chooses a target class as a pollution source to generate a perturbed image and tricks the classifier into classifying it as the target class. However, the effectiveness of adversarial attacks suffers from some limitations, including the need for specific control over the labeling of the target class and challenges in generating effective adversarial samples in real-world scenarios. To overcome these limitations, we propose a novel clean-label attack based on negative afterimage. Negative afterimage is a visual phenomenon whereby, after gazing at a bright image for some time, one perceives the inverse color image of the previously observed bright region when shifting focus to a darker area. We explore the afterimage phenomenon and use negative afterimages as pollution sources to generate adversarial samples, which can avoid any dependency on the labeling of pollution sources. Subsequently, we generate adversarial samples using an optimization-based approach to ensure that adversarial samples are visually undetectable. We conducted experiments on two widely used datasets, CIFAR-10 and ImageNet. The results show that our proposed method can effectively generate adversarial samples with high stealthiness.
引用
收藏
页码:449 / 460
页数:12
相关论文
共 50 条
  • [21] Projective Ranking: A Transferable Evasion Attack Method on Graph Neural Networks
    Zhang, He
    Wu, Bang
    Yang, Xiangwen
    Zhou, Chuan
    Wang, Shuo
    Yuan, Xingliang
    Pan, Shirui
    [J]. PROCEEDINGS OF THE 30TH ACM INTERNATIONAL CONFERENCE ON INFORMATION & KNOWLEDGE MANAGEMENT, CIKM 2021, 2021, : 3617 - 3621
  • [22] Diversity Adversarial Training against Adversarial Attack on Deep Neural Networks
    Kwon, Hyun
    Lee, Jun
    [J]. SYMMETRY-BASEL, 2021, 13 (03):
  • [23] Exploration of Membership Inference Attack on Convolutional Neural Networks and Its Defenses
    Yao, Yimian
    [J]. 2022 INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, COMPUTER VISION AND MACHINE LEARNING (ICICML), 2022, : 604 - 610
  • [24] Label-Only Membership Inference Attack Based on Model Explanation
    Ma, Yao
    Zhai, Xurong
    Yu, Dan
    Yang, Yuli
    Wei, Xingyu
    Chen, Yongle
    [J]. NEURAL PROCESSING LETTERS, 2024, 56 (05)
  • [25] Synergy between traditional classification and classification based on negative features in deep convolutional neural networks
    Milosevic, Nemanja
    Rackovic, Milos
    [J]. NEURAL COMPUTING & APPLICATIONS, 2021, 33 (13) : 7593 - 7602
  • [26] Synergy between traditional classification and classification based on negative features in deep convolutional neural networks
    Nemanja Milošević
    Miloš Racković
    [J]. Neural Computing and Applications, 2021, 33 : 7593 - 7602
  • [27] Neural Network Based Attack on a Masked Implementation of AES
    Gilmore, Richard
    Hanley, Neil
    O'Neill, Maire
    [J]. 2015 IEEE INTERNATIONAL SYMPOSIUM ON HARDWARE ORIENTED SECURITY AND TRUST (HOST), 2015, : 106 - 111
  • [28] Dual-Targeted adversarial example in evasion attack on graph neural networks
    Kwon, Hyun
    Kim, Dae-Jin
    [J]. SCIENTIFIC REPORTS, 2025, 15 (01):
  • [29] Correlation-Aware Neural Networks for DDoS Attack Detection in IoT Systems
    Hekmati, Arvin
    Zhang, Jiahe
    Sarkar, Tamoghna
    Jethwa, Nishant
    Grippo, Eugenio
    Krishnamachari, Bhaskar
    [J]. IEEE-ACM TRANSACTIONS ON NETWORKING, 2024, 32 (05) : 3929 - 3944
  • [30] Neural Networks for DDoS Attack Detection using an Enhanced Urban IoT Dataset
    Hekmati, Arvin
    Grippo, Eugenio
    Krishnamachari, Bhaskar
    [J]. 2022 31ST INTERNATIONAL CONFERENCE ON COMPUTER COMMUNICATIONS AND NETWORKS (ICCCN 2022), 2022,