mlsanityData to 7 Oct 2026

SAIF: Sparse Adversarial and Imperceptible Attack Framework

arXiv 2212.07495 ↗First posted 14 Dec 2022cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

Adversarial attacks hamper the decision-making ability of neural networks by perturbing the input signal. The addition of calculated small distortion to images, for instance, can deceive a well-trained image classification network. In this work, we propose a novel attack technique called Sparse Adversarial and Interpretable Attack Framework (SAIF). Specifically, we design imperceptible attacks that contain low-magnitude perturbations at a small number of pixels and leverage these sparse attacks to reveal the vulnerability of classifiers. We use the Frank-Wolfe (conditional gradient) algorithm to simultaneously optimize the attack perturbations for bounded magnitude and sparsity with O(1/√T) convergence. Empirical results show that SAIF computes highly imperceptible and interpretable adversarial examples, and outperforms state-of-the-art sparse attack methods on the ImageNet dataset.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 29 of its 43 references that are in the map. First posted on arXiv 14 Dec 2022.

Builds on

29 of 43 references.

References · 29

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2013Intriguing properties of neural networkscited by 5,114
  6. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  7. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  8. 2015Fast R-CNNcited by 2,930
  9. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  10. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  11. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  12. 2017Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networkscited by 936
17 more
  1. 2015The Limitations of Deep Learning in Adversarial Settingscited by 784
  2. 2017One pixel attack for fooling deep neural networkscited by 579
  3. 2021On the Convergence and Robustness of Adversarial Trainingcited by 127
  4. 2019Sparse and Imperceivable Adversarial Attackscited by 95
  5. 2018SparseFool: a few pixels make a big differencecited by 94
  6. 2018Structured Adversarial Attack: Towards General Implementation and Better Interpretabilitycited by 63
  7. 2016Convergence Rate of Frank-Wolfe for Non-Convex Objectivescited by 59
  8. 2020GreedyFool: Distortion-Aware Sparse Adversarial Attackcited by 37
  9. 2018A Frank-Wolfe Framework for Efficient and Effective Adversarial Attackscited by 29
  10. 2020Double Targeted Universal Adversarial Perturbationscited by 23
  11. 2021Robust Models Are More Interpretable Because Attributions Look Normalcited by 21
  12. 2021Sparse and Imperceptible Adversarial Attack via a Homotopy Algorithmcited by 15
  13. 2020Understanding and Increasing Efficiency of Frank-Wolfe Adversarial Trainingcited by 14
  14. 2021Transferable Sparse Adversarial Attackcited by 9
  15. 2021FrankWolfe.jl: a high-performance and flexible toolbox for Frank-Wolfe algorithms and Conditional Gradientscited by 8
  16. 2018Frank-Wolfe Splitting via Augmented Lagrangian Methodcited by 7
  17. 2021Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance Orderingscited by 5

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025ZQBA: Zero Query Black-box Adversarial Attackcited by 1