mlsanityData to 7 Oct 2026

A Unified Game-Theoretic Interpretation of Adversarial Robustness

arXiv 2103.07364 ↗First posted 12 Mar 2021cs.LG cs.AI cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

This paper provides a unified view to explain different adversarial attacks and defense methods, i.e. the view of multi-order interactions between input variables of DNNs. Based on the multi-order interaction, we discover that adversarial attacks mainly affect high-order interactions to fool the DNN. Furthermore, we find that the robustness of adversarially trained DNNs comes from category-specific low-order interactions. Our findings provide a potential method to unify adversarial perturbations and robustness, which can explain the existing defense methods in a principle way. Besides, our findings also make a revision of previous inaccurate understanding of the shape bias of adversarially learned features.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 64 of its 87 references that are in the map. First posted on arXiv 12 Mar 2021.

Builds on

64 of 87 references.

References · 64

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2016Densely Connected Convolutional Networkscited by 6,691
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  7. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  8. 2013Intriguing properties of neural networkscited by 5,114
  9. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  10. 2016Wide Residual Networkscited by 3,057
  11. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  12. 2016Adversarial examples in the physical worldcited by 1,762
38 more
  1. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  2. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  3. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  4. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  5. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 950
  6. 2016Practical Black-Box Attacks against Machine Learningcited by 893
  7. 2018Robustness May Be at Odds with Accuracycited by 873
  8. 2015The Limitations of Deep Learning in Adversarial Settingscited by 784
  9. 2017ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Modelscited by 656
  10. 2016Delving into Transferable Adversarial Examples and Black-box Attackscited by 530
  11. 2018Black-box Adversarial Attacks with Limited Queries and Informationcited by 499
  12. 2020Do Adversarially Robust ImageNet Models Transfer Better?cited by 315
  13. 2018Feature Denoising for Improving Adversarial Robustnesscited by 312
  14. 2019High Frequency Component Helps Explain the Generalization of Convolutional Neural Networkscited by 310
  15. 2019A Fourier Perspective on Model Robustness in Computer Visioncited by 288
  16. 2018Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionalitycited by 280
  17. 2018Consistent Individualized Feature Attribution for Tree Ensemblescited by 279
  18. 2017MagNet: a Two-Pronged Defense against Adversarial Examplescited by 268
  19. 2017Formal Guarantees on the Robustness of a Classifier against Adversarial Manipulationcited by 182
  20. 2018Evaluating the Robustness of Neural Networks: An Extreme Value Theory Approachcited by 181
  21. 2020Skip Connections Matter: On the Transferability of Adversarial Examples Generated with ResNetscited by 165
  22. 2019The Shapley Taylor Interaction Indexcited by 143
  23. 2018Constructing Unrestricted Adversarial Examples with Generative Modelscited by 135
  24. 2019Explaining Deep Neural Networks with a Polynomial Time Algorithm for Shapley Values Approximationcited by 115
  25. 2020Explaining Explanations: Axiomatic Feature Interactions for Deep Networkscited by 103
  26. 2017Keeping the Bad Guys Out: Protecting and Vaccinating Deep Learning with JPEG Compressioncited by 86
  27. 2017Detecting Statistical Interactions from Neural Network Weightscited by 80
  28. 2015Analysis of classifiers' robustness to adversarial perturbationscited by 76
  29. 2021Improving Adversarial Robustness via Channel-wise Activation Suppressingcited by 76
  30. 2019Interpreting Adversarially Trained Convolutional Neural Networkscited by 75
  31. 2020A Unified Approach to Interpreting and Boosting Adversarial Transferabilitycited by 71
  32. 2018Beyond Word Importance: Contextual Decomposition to Extract Interactions from LSTMscited by 67
  33. 2018Hierarchical interpretations for neural network predictionscited by 66
  34. 2018Structured Adversarial Attack: Towards General Implementation and Better Interpretabilitycited by 63
  35. 2017Houdini: Fooling Deep Structured Prediction Modelscited by 50
  36. 2019ML-LOO: Detecting Adversarial Examples with Feature Attributioncited by 48
  37. 2020Proper Network Interpretability Helps Adversarial Robustness in Classificationcited by 41
  38. 2017Analyzing the Robustness of Nearest Neighbors to Adversarial Examplescited by 36

50 most cited of 64 All 64 (JSON)

Built on by

Not cited yet.