mlsanityData to 7 Oct 2026

Impact of Spatial Frequency Based Constraints on Adversarial Robustness

arXiv 2104.12679 ↗First posted 26 Apr 2021cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

Adversarial examples mainly exploit changes to input pixels to which humans are not sensitive to, and arise from the fact that models make decisions based on uninterpretable features. Interestingly, cognitive science reports that the process of interpretability for human classification decision relies predominantly on low spatial frequency components. In this paper, we investigate the robustness to adversarial perturbations of models enforced during training to leverage information corresponding to different spatial frequency ranges. We show that it is tightly linked to the spatial frequency characteristics of the data at stake. Indeed, depending on the data set, the same constraint may results in very different level of robustness (up to 0.41 adversarial accuracy difference). To explain this phenomenon, we conduct several experiments to enlighten influential factors such as the level of sensitivity to high frequencies, and the transferability of adversarial perturbations between original and low-pass filtered inputs.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 29 of its 37 references that are in the map. First posted on arXiv 26 Apr 2021.

Builds on

29 of 37 references.

References · 29

  1. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2013Intriguing properties of neural networkscited by 5,114
  4. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  5. 2016Wide Residual Networkscited by 3,057
  6. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  7. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
  8. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  9. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 950
  10. 2018Robustness May Be at Odds with Accuracycited by 873
  11. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 591
  12. 2020Overfitting in adversarially robust deep learningcited by 504
17 more
  1. 2019On Evaluating Adversarial Robustnesscited by 442
  2. 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
  3. 2019High Frequency Component Helps Explain the Generalization of Convolutional Neural Networkscited by 310
  4. 2019A Fourier Perspective on Model Robustness in Computer Visioncited by 288
  5. 2017MagNet: a Two-Pronged Defense against Adversarial Examplescited by 268
  6. 2018Feature Distillation: DNN-Oriented JPEG Compression Against Adversarial Examplescited by 98
  7. 2017Measuring the tendency of CNNs to Learn Surface Statistical Regularitiescited by 87
  8. 2019On the Connection Between Adversarial Robustness and Saliency Map Interpretabilitycited by 80
  9. 2019Interpreting Adversarially Trained Convolutional Neural Networkscited by 75
  10. 2018Shield: Fast, Practical Defense and Vaccination for Deep Learning using JPEG Compressioncited by 74
  11. 2019On the Effectiveness of Low Frequency Perturbationscited by 56
  12. 2019Are Perceptually-Aligned Gradients a General Property of Robust Classifiers?cited by 42
  13. 2019Jacobian Adversarially Regularized Networks for Robustnesscited by 40
  14. 2020Towards Frequency-Based Explanation for Robust CNNcited by 29
  15. 2019PuVAE: A Variational Autoencoder to Purify Adversarial Examplescited by 28
  16. 2019Adversarial Defense by Suppressing High-frequency Componentscited by 22
  17. 2020Towards Achieving Adversarial Robustness by Enforcing Feature Consistency Across Bit Planescited by 15

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 1; 2023: 5; 2024: 1.

8 papers.

  1. 2021A Frequency Perspective of Adversarial Robustnesscited by 28
  2. 2023Spatial-frequency channels, shape bias, and adversarial robustnesscited by 28
  3. 2023Towards Building More Robust Models with Frequency Biascited by 12
  4. 2023An Extended Study of Human-like Behavior under Adversarial Trainingcited by 9
  5. 2023Fourier Analysis on Robustness of Graph Convolutional Neural Networks for Skeleton-based Action Recognitioncited by 2
  6. 2024DFT-Based Adversarial Attack Detection in MRI Brain Imaging: Enhancing Diagnostic Accuracy in Alzheimer's Case Studiescited by 1
  7. 2022Testing predictions of representation cost theory with CNNscited by 0
  8. 2023A Spectral Perspective towards Understanding and Improving Adversarial Robustnesscited by 0