mlsanityData to 7 Oct 2026

Meta Gradient Adversarial Attack

arXiv 2108.04204 ↗First posted 9 Aug 2021cs.CVcited by 55

Not on the map. Why?

Abstract (arXiv)

In recent years, research on adversarial attacks has become a hot spot. Although current literature on the transfer-based adversarial attack has achieved promising results for improving the transferability to unseen black-box models, it still leaves a long way to go. Inspired by the idea of meta-learning, this paper proposes a novel architecture called Meta Gradient Adversarial Attack (MGAA), which is plug-and-play and can be integrated with any existing gradient-based attack method for improving the cross-model transferability. Specifically, we randomly sample multiple models from a model zoo to compose different tasks and iteratively simulate a white-box attack and a black-box attack in each task. By narrowing the gap between the gradient directions in white-box and black-box attacks, the transferability of adversarial examples on the black-box setting can be improved. Extensive experiments on the CIFAR10 and ImageNet datasets show that our architecture outperforms the state-of-the-art methods for both black-box and white-box attack settings.

Key facts. Cited by 55 papers in the map (data through 7 Oct 2026). Builds on 50 of its 54 references that are in the map. First posted on arXiv 9 Aug 2021.

Builds on

50 of 54 references.

References · 50

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  6. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  7. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  8. 2017Squeeze-and-Excitation Networkscited by 4,763
  9. 2014Going Deeper with Convolutionscited by 4,231
  10. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  11. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
  12. 2018CBAM: Convolutional Block Attention Modulecited by 2,463
38 more
  1. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  2. 2016Adversarial examples in the physical worldcited by 1,762
  3. 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,509
  4. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  5. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  6. 2017Learning Transferable Architectures for Scalable Image Recognitioncited by 1,097
  7. 2018On First-Order Meta-Learning Algorithmscited by 1,003
  8. 2020Designing Network Design Spacescited by 954
  9. 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
  10. 2017Learning to Generalize: Meta-Learning for Domain Generalizationcited by 762
  11. 2017ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Modelscited by 656
  12. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 591
  13. 2016Delving into Transferable Adversarial Examples and Black-box Attackscited by 530
  14. 2017Progressive Neural Architecture Searchcited by 483
  15. 2017Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Modelscited by 459
  16. 2019Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attackscited by 440
  17. 2017Deep Layer Aggregationcited by 436
  18. 2017Mitigating Adversarial Effects Through Randomizationcited by 382
  19. 2018Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Modelscited by 362
  20. 2019Nesterov Accelerated Gradient and Scale Invariance for Adversarial Attackscited by 352
  21. 2017Defense against Adversarial Attacks Using High-Level Representation Guided Denoisercited by 329
  22. 2018Certified Defenses against Adversarial Examplescited by 305
  23. 2019Improving Adversarial Robustness via Promoting Ensemble Diversitycited by 213
  24. 2016A study of the effect of JPG compression on adversarial imagescited by 196
  25. 2019Efficient Decision-based Black-box Adversarial Attacks on Face Recognitioncited by 173
  26. 2018Stochastic Activation Pruning for Robust Adversarial Defensecited by 171
  27. 2016Deep Pyramidal Residual Networkscited by 163
  28. 2017Towards Robust Neural Networks via Random Self-ensemblecited by 158
  29. 2018Scaling provable adversarial defensescited by 141
  30. 2018Query-Efficient Hard-label Black-box Attack:An Optimization-based Approachcited by 130
  31. 2019Improving Black-box Adversarial Attacks with a Transfer-based Priorcited by 121
  32. 2018Adversarial Attacks and Defences Competitioncited by 96
  33. 2019NATTACK: Learning the Distributions of Adversarial Examples for an Improved Black-Box Attack on Deep Neural Networkscited by 94
  34. 2017Shake-Shake regularizationcited by 87
  35. 2019Enhancing Adversarial Defense by k-Winners-Take-Allcited by 51
  36. 2018Are Generative Classifiers More Robust to Adversarial Attacks?cited by 36
  37. 2019Query-efficient Meta Attack to Deep Neural Networkscited by 33
  38. 2019Robust Local Features for Improving the Generalization of Adversarial Trainingcited by 30

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 8; 2023: 17; 2024: 14; 2025: 7; 2026: 4.

55 papers.

Sort:

By area:

  1. 2023To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Nowcited by 204
  2. 2022Physical Adversarial Attack meets Computer Vision: A Decade Surveycited by 66
  3. 2023How Deep Learning Sees the World: A Survey on Adversarial Attacks & Defensescited by 65
  4. 2024R.A.C.E.: Robust Adversarial Concept Erasure for Secure Text-to-Image Diffusion Modelcited by 56
  5. 2021Harnessing Perceptual Adversarial Patches for Crowd Countingcited by 39
  6. 2022How to Robustify Black-Box ML Models? A Zeroth-Order Optimization Perspectivecited by 39
  7. 2023Generalizable Black-Box Adversarial Attack with Meta Learningcited by 29
  8. 2024Towards Robust Physical-world Backdoor Attacks on Lane Detectioncited by 25
  9. 2023Physical Adversarial Attacks For Camera-based Smart Systems: Current Trends, Categorization, Applications, Research Challenges, and Future Outlookcited by 20
  10. 2022Transferable Graph Backdoor Attackcited by 19
  11. 2023Attacks in Adversarial Machine Learning: A Systematic Survey from the Life-cycle Perspectivecited by 19
  12. 2023Ensemble-based Blackbox Attacks on Dense Predictioncited by 16
38 more
  1. 2021Training Meta-Surrogate Model for Transferable Adversarial Attackcited by 15
  2. 2024Mitigating the Curse of Dimensionality for Certified Robustness via Dual Randomized Smoothingcited by 15
  3. 2023Mutual-modality Adversarial Attack with Semantic Perturbationcited by 14
  4. 2021Learning to Learn Transferable Attackcited by 13
  5. 2022Blackbox Attacks via Surrogate Ensemble Searchcited by 13
  6. 2024Machine Learning Robustness: A Primercited by 8
  7. 2023Evading Forensic Classifiers with Attribute-Conditioned Adversarial Facescited by 6
  8. 2023LEAP: Efficient and Automated Test Method for NLP Softwarecited by 6
  9. 2023SoK: Pitfalls in Evaluating Black-Box Attackscited by 6
  10. 2024Certification of Speaker Recognition Models to Additive Perturbationscited by 6
  11. 2022Sampling-based Fast Gradient Rescaling Method for Highly Transferable Adversarial Attackscited by 5
  12. 2022Perturbation Inactivation Based Adversarial Defense for Face Recognitioncited by 5
  13. 2023Meta Generative Attack on Person Reidentificationcited by 5
  14. 2023DOEPatch: Dynamically Optimized Ensemble Model for Adversarial Patches Generationcited by 5
  15. 2024Meta Invariance Defense Towards Generalizable Robustness to Unknown Adversarial Attackscited by 5
  16. 2024Catastrophic Overfitting: A Potential Blessing in Disguisecited by 4
  17. 2022Adversarial Attack via Dual-Stage Network Erosioncited by 3
  18. 2024Exploring Frequencies via Feature Mixing and Meta-Learning for Improving Adversarial Transferabilitycited by 3
  19. 2025Defensive Adversarial CAPTCHA: A Semantics-Driven Framework for Natural Adversarial Example Generationcited by 3
  20. 2021Adversarial Attack across Datasetscited by 2
  21. 2023RecUP-FL: Reconciling Utility and Privacy in Federated Learning via User-configurable Privacy Defensecited by 2
  22. 2024Advancing Generalized Transfer Attack with Initialization Derived Bilevel Optimization and Dynamic Sequence Truncationcited by 2
  23. 2024Artificial Immune System of Secure Face Recognition Against Adversarial Attackscited by 2
  24. 2024Take Fake as Real: Realistic-like Robust Black-box Adversarial Attack to Evade AIGC Detectioncited by 2
  25. 2025Web Artifact Attacks Disrupt Vision Language Modelscited by 2
  26. 2025Text Adversarial Attacks with Dynamic Outputscited by 2
  27. 2025Boosting Adversarial Transferability via Ensemble Non-Attentioncited by 2
  28. 2025Angular Gradient Sign Method: Uncovering Vulnerabilities in Hyperbolic Networkscited by 2
  29. 2021Meta Adversarial Perturbationscited by 1
  30. 2024Sparse and Transferable Universal Singular Vectors Attackcited by 1
  31. 2024Data Poisoning: An Overlooked Threat to Power Grid Resiliencecited by 1
  32. 2024Learning to Learn Transferable Generative Attack for Person Re-Identificationcited by 1
  33. 2025Towards Adaptive Meta-Gradient Adversarial Examples for Visual Trackingcited by 1
  34. 2025Multi-task Adversarial Attacks against Black-box Model with Few-shot Queriescited by 1
  35. 2026Certified vs. Empirical Adversarial Robust-ness via Hybrid Convolutions with Attention Stochasticitycited by 1
  36. 2022Gradient Concealment: Free Lunch for Defending Adversarial Attackscited by 0
  37. 2023Interpretable Spectrum Transformation Attacks to Speaker Recognitioncited by 0
  38. 2023AF2-Mutation: Adversarial Sequence Mutations against AlphaFold2 on Protein Tertiary Structure Predictioncited by 0

50 most cited of 55 All 55 (JSON)