mlsanityData to 14 Oct 2026

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

arXiv 2212.05827 ↗First posted 12 Dec 2022cs.CV cs.CR cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 31 of its 40 references that are in the map. First posted on arXiv 12 Dec 2022.

Builds on

31 of 40 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  5. 2013Intriguing properties of neural networkscited by 5,155
  6. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,859
  7. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
  8. 2016YOLO9000: Better, Faster, Strongercited by 1,431
  9. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,340
  10. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 953
  11. 2016Adversarial Machine Learning at Scalecited by 936
  12. 2016Universal adversarial perturbationscited by 784
19 more
  1. 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 776
  2. 2017Evasion Attacks against Machine Learning at Test Timecited by 689
  3. 2018BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentationcited by 645
  4. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 599
  5. 2020On Adaptive Attacks to Adversarial Example Defensescited by 555
  6. 2016Delving into Transferable Adversarial Examples and Black-box Attackscited by 536
  7. 2017Adversarial Patchcited by 517
  8. 2018Black-box Adversarial Attacks with Limited Queries and Informationcited by 501
  9. 2016Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samplescited by 446
  10. 2019Fooling automated surveillance cameras: adversarial patches to attack person detectioncited by 268
  11. 2018Adversarial Risk and the Dangers of Evaluating Against Weak Attackscited by 221
  12. 2018Physical Adversarial Examples for Object Detectorscited by 163
  13. 2018DPatch: An Adversarial Patch Attack on Object Detectorscited by 147
  14. 2019On Physical Adversarial Patches for Object Detectioncited by 93
  15. 2020Perturbing Across the Feature Hierarchy to Improve Standard and Strict Blackbox Attack Transferabilitycited by 57
  16. 2021Evaluating the Robustness of Semantic Segmentation for Autonomous Driving against Real-World Adversarial Patch Attackscited by 55
  17. 2019Role of Spatial Context in Adversarial Robustness for Object Detectioncited by 35
  18. 2022Give Me Your Attention: Dot-Product Attention Considered Harmful for Adversarial Patch Robustnesscited by 23
  19. 2016LOTS about Attacking Deep Featurescited by 14

Built on by

Not cited yet.