mlsanityData to 7 Oct 2026

Meta Adversarial Perturbations

arXiv 2111.10291 ↗First posted 19 Nov 2021cs.LG cs.AI cs.CRcited by 1

Not on the map. Why?

Abstract (arXiv)

A plethora of attack methods have been proposed to generate adversarial examples, among which the iterative methods have been demonstrated the ability to find a strong attack. However, the computation of an adversarial perturbation for a new data point requires solving a time-consuming optimization problem from scratch. To generate a stronger attack, it normally requires updating a data point with more iterations. In this paper, we show the existence of a meta adversarial perturbation (MAP), a better initialization that causes natural images to be misclassified with high probability after being updated through only a one-step gradient ascent update, and propose an algorithm for computing such perturbations. We conduct extensive experiments, and the empirical results demonstrate that state-of-the-art deep neural networks are vulnerable to meta perturbations. We further show that these perturbations are not only image-agnostic, but also model-agnostic, as a single perturbation generalizes well across unseen data points and different neural network architectures.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 31 of its 37 references that are in the map. First posted on arXiv 19 Nov 2021.

Builds on

31 of 37 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2016Densely Connected Convolutional Networkscited by 6,691
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2013Intriguing properties of neural networkscited by 5,114
  7. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  8. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  9. 2017Squeeze-and-Excitation Networkscited by 4,763
  10. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  11. 2016Adversarial examples in the physical worldcited by 1,762
  12. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
19 more
  1. 2018On First-Order Meta-Learning Algorithmscited by 1,003
  2. 2016Universal adversarial perturbationscited by 774
  3. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  4. 2016Learning to learn by gradient descent by gradient descentcited by 616
  5. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 591
  6. 2019Meta-Learning with Implicit Gradientscited by 523
  7. 2018Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Surveycited by 459
  8. 2017Wild Patterns: Ten Years After the Rise of Adversarial Machine Learningcited by 411
  9. 2017Meta-SGD: Learning to Learn Quickly for Few-Shot Learningcited by 377
  10. 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
  11. 2018Adversarial Attacks and Defences: A Surveycited by 259
  12. 2016Learning to Optimizecited by 195
  13. 2019You Only Propagate Once: Accelerating Adversarial Training via Maximal Principlecited by 181
  14. 2017EAD: Elastic-Net Attacks to Deep Neural Networks via Adversarial Examplescited by 154
  15. 2021Meta Gradient Adversarial Attackcited by 55
  16. 2019Efficient Adversarial Training with Transferable Adversarial Examplescited by 54
  17. 2019Alpha MAML: Adaptive Model-Agnostic Meta-Learningcited by 25
  18. 2018Deep Meta-Learning: Learning to Learn in the Concept Spacecited by 22
  19. 2021Model-Agnostic Meta-Attack: Towards Reliable Evaluation of Adversarial Robustnesscited by 3

Built on by

Cited by year

Citing papers per year: 2022: 1.

1 paper.

  1. 2022Boosting Black-Box Adversarial Attacks with Meta Learningcited by 1