mlsanityData to 7 Oct 2026

Diversified Adversarial Attacks based on Conjugate Gradient Method

arXiv 2206.09628 ↗First posted 20 Jun 2022cs.LG cs.CR cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

Deep learning models are vulnerable to adversarial examples, and adversarial attacks used to generate such examples have attracted considerable research interest. Although existing methods based on the steepest descent have achieved high attack success rates, ill-conditioned problems occasionally reduce their performance. To address this limitation, we utilize the conjugate gradient (CG) method, which is effective for this type of problem, and propose a novel attack algorithm inspired by the CG method, named the Auto Conjugate Gradient (ACG) attack. The results of large-scale evaluation experiments conducted on the latest robust models show that, for most models, ACG was able to find more adversarial examples with fewer iterations than the existing SOTA algorithm Auto-PGD (APGD). We investigated the difference in search performance between ACG and APGD in terms of diversification and intensification, and define a measure called Diversity Index (DI) to quantify the degree of diversity. From the analysis of the diversity using this index, we show that the more diverse search of the proposed method remarkably improves its attack success rate.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 36 of its 53 references that are in the map. First posted on arXiv 20 Jun 2022.

Builds on

36 of 53 references.

References · 36

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2013Intriguing properties of neural networkscited by 5,114
  4. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  5. 2016Adversarial examples in the physical worldcited by 1,762
  6. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  7. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  8. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  9. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 682
  10. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 659
  11. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 591
  12. 2020Overfitting in adversarially robust deep learningcited by 504
24 more
  1. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  2. 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
  3. 2019Nesterov Accelerated Gradient and Scale Invariance for Adversarial Attackscited by 352
  4. 2019Minimally distorted Adversarial Examples with a Fast Adaptive Boundary Attackcited by 333
  5. 2020Do Adversarially Robust ImageNet Models Transfer Better?cited by 315
  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. 2021Improving Robustness using Generated Datacited by 233
  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. 2020Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuningcited by 152
  14. 2020Self-Adaptive Training: beyond Empirical Risk Minimizationcited by 126
  15. 2021Robust Learning Meets Generative Models: Can Proxy Distributions Improve Adversarial Robustness?cited by 113
  16. 2020Boosting Adversarial Training with Hypersphere Embeddingcited by 97
  17. 2020Learnable Boundary Guided Adversarial Trainingcited by 93
  18. 2021Exploring Architectural Ingredients of Adversarially Robust Deep Neural Networkscited by 73
  19. 2019An Alternative Surrogate Loss for PGD-based Adversarial Testingcited by 45
  20. 2020Efficient Robust Training via Backward Smoothingcited by 29
  21. 2020SAT: Improving Adversarial Training via Curriculum-Based Loss Smoothingcited by 26
  22. 2018Distributionally Adversarial Attackcited by 24
  23. 2021Improving Neural Network Robustness via Persistency of Excitationcited by 14
  24. 2018Trust Region Based Adversarial Attack on Neural Networkscited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 2; 2024: 5; 2025: 2; 2026: 1.

11 papers.

  1. 2024Privacy-preserving Universal Adversarial Defense for Black-box Modelscited by 3
  2. 2024Towards Million-Scale Adversarial Robustness Evaluation With Stronger Individual Attackscited by 3
  3. 2022Alternating Objectives Generates Stronger PGD-Based Adversarial Attackscited by 2
  4. 2025On the Interaction of Compressibility and Adversarial Robustnesscited by 2
  5. 2024SoK: Analyzing Adversarial Examples: A Framework to Study Adversary Knowledgecited by 1
  6. 2025MOS-Attack: A Scalable Multi-objective Adversarial Attack Frameworkcited by 1
  7. 2026NetDiffuser: Deceiving DNN-Based Network Attack Detection Systems with Diffusion-Generated Adversarial Trafficcited by 1
  8. 2023Interpretable Spectrum Transformation Attacks to Speaker Recognitioncited by 0
  9. 2023RobustFair: Adversarial Evaluation through Fairness Confusion Directed Gradient Searchcited by 0
  10. 2024A Novel Bifurcation Method for Observation Perturbation Attacks on Reinforcement Learning Agents: Load Altering Attacks on a Cyber Physical Power Systemcited by 0
  11. 2024Enhancing Output Diversity Improves Conjugate Gradient-based Adversarial Attackscited by 0