mlsanityData to 7 Oct 2026

Rethinking Classifier and Adversarial Attack

arXiv 2205.02743 ↗First posted 4 May 2022cs.LG cs.AI cs.CRcited by 0

Not on the map. Why?

Abstract (arXiv)

Various defense models have been proposed to resist adversarial attack algorithms, but existing adversarial robustness evaluation methods always overestimate the adversarial robustness of these models (i.e., not approaching the lower bound of robustness). To solve this problem, this paper uses the proposed decouple space method to divide the classifier into two parts: non-linear and linear. Then, this paper defines the representation vector of the original example (and its space, i.e., the representation space) and uses the iterative optimization of Absolute Classification Boundaries Initialization (ACBI) to obtain a better attack starting point. Particularly, this paper applies ACBI to nearly 50 widely-used defense models (including 8 architectures). Experimental results show that ACBI achieves lower robust accuracy in all cases.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 46 references that are in the map. First posted on arXiv 4 May 2022.

Builds on

32 of 46 references.

References · 32

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  4. 2016Densely Connected Convolutional Networkscited by 6,691
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  7. 2013Intriguing properties of neural networkscited by 5,114
  8. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  9. 2016Adversarial examples in the physical worldcited by 1,762
  10. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  11. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  12. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
20 more
  1. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  2. 2020Overfitting in adversarially robust deep learningcited by 504
  3. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  4. 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
  5. 2019Minimally distorted Adversarial Examples with a Fast Adaptive Boundary Attackcited by 333
  6. 2020Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examplescited by 275
  7. 2020Attacks Which Do Not Kill Training Make Adversarial Learning Strongercited by 242
  8. 2020Understanding and Improving Fast Adversarial Trainingcited by 213
  9. 2021Fixing Data Augmentation to Improve Adversarial Robustnesscited by 201
  10. 2020Geometry-aware Instance-reweighted Adversarial Trainingcited by 198
  11. 2019You Only Propagate Once: Accelerating Adversarial Training via Maximal Principlecited by 181
  12. 2018MMA Training: Direct Input Space Margin Maximization through Adversarial Trainingcited by 171
  13. 2020Self-Adaptive Training: beyond Empirical Risk Minimizationcited by 126
  14. 2020Boosting Adversarial Training with Hypersphere Embeddingcited by 97
  15. 2019Defense Against Adversarial Attacks Using Feature Scattering-based Adversarial Trainingcited by 95
  16. 2020Learnable Boundary Guided Adversarial Trainingcited by 93
  17. 2019Controlling Neural Level Setscited by 76
  18. 2022Practical Evaluation of Adversarial Robustness via Adaptive Auto Attackcited by 44
  19. 2020A Tunable Robust Pruning Framework Through Dynamic Network Rewiring of DNNscited by 37
  20. 2020Manifold Regularization for Locally Stable Deep Neural Networkscited by 8

Built on by

Not cited yet.