mlsanityData to 7 Oct 2026

Understanding Adversarial Robustness from Feature Maps of Convolutional Layers

arXiv 2202.12435 ↗First posted 25 Feb 2022cs.CV cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

The adversarial robustness of a neural network mainly relies on two factors: model capacity and anti-perturbation ability. In this paper, we study the anti-perturbation ability of the network from the feature maps of convolutional layers. Our theoretical analysis discovers that larger convolutional feature maps before average pooling can contribute to better resistance to perturbations, but the conclusion is not true for max pooling. It brings new inspiration to the design of robust neural networks and urges us to apply these findings to improve existing architectures. The proposed modifications are very simple and only require upsampling the inputs or slightly modifying the stride configurations of downsampling operators. We verify our approaches on several benchmark neural network architectures, including AlexNet, VGG, RestNet18, and PreActResNet18. Non-trivial improvements in terms of both natural accuracy and adversarial robustness can be achieved under various attack and defense mechanisms. The code is available at urlhttps://github.com/MTandHJ/rcm.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 34 of its 62 references that are in the map. First posted on arXiv 25 Feb 2022.

Builds on

34 of 62 references.

References · 34

  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. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2013Intriguing properties of neural networkscited by 5,114
  7. 2016Wide Residual Networkscited by 3,057
  8. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  9. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  10. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  11. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  12. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
22 more
  1. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  2. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  3. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  4. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 950
  5. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  6. 2020Overfitting in adversarially robust deep learningcited by 504
  7. 2019Provably Robust Deep Learning via Adversarially Trained Smoothed Classifierscited by 328
  8. 2020Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examplescited by 275
  9. 2020Attacks Which Do Not Kill Training Make Adversarial Learning Strongercited by 242
  10. 2018Adversarial Logit Pairingcited by 224
  11. 2019Generalization Bounds of Stochastic Gradient Descent for Wide and Deep Neural Networkscited by 221
  12. 2021Fixing Data Augmentation to Improve Adversarial Robustnesscited by 201
  13. 2020Bag of Tricks for Adversarial Trainingcited by 199
  14. 2019Adversarial Training and Robustness for Multiple Perturbationscited by 182
  15. 2020A Self-supervised Approach for Adversarial Robustnesscited by 180
  16. 2021On the Adversarial Robustness of Vision Transformerscited by 104
  17. 2020Do Wider Neural Networks Really Help Adversarial Robustness?cited by 79
  18. 2021Exploring Architectural Ingredients of Adversarially Robust Deep Neural Networkscited by 73
  19. 2021Parameterizing Activation Functions for Adversarial Robustnesscited by 22
  20. 2017Foolbox: A Python toolbox to benchmark the robustness of machine learning modelscited by 19
  21. 2019Training Provably Robust Models by Polyhedral Envelope Regularizationcited by 6
  22. 2021Adversarial Information Bottleneckcited by 5

Built on by

Not cited yet.