Carpet-bombing patch: attacking a deep network without usual requirements

arxiv(2022)

引用 0|浏览2
暂无评分
摘要
Although deep networks have shown vulnerability to evasion attacks, such attacks have usually unrealistic requirements. Recent literature discussed the possibility to remove or not some of these requirements. This paper contributes to this literature by introducing a carpet-bombing patch attack which has almost no requirement. Targeting the feature representations, this patch attack does not require knowing the network task. This attack decreases accuracy on Imagenet, mAP on Pascal Voc, and IoU on Cityscapes without being aware that the underlying tasks involved classification, detection or semantic segmentation, respectively. Beyond the potential safety issues raised by this attack, the impact of the carpet-bombing attack highlights some interesting property of deep network layer dynamic.
更多
查看译文
关键词
deep network,patch,carpet-bombing
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要