mlsanityData to 14 Oct 2026

GradMDM: Adversarial Attack on Dynamic Networks

arXiv 2304.06724 ↗First posted 1 Apr 2023cs.CR cs.CV cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

Dynamic neural networks can greatly reduce computation redundancy without compromising accuracy by adapting their structures based on the input. In this paper, we explore the robustness of dynamic neural networks against energy-oriented attacks targeted at reducing their efficiency. Specifically, we attack dynamic models with our novel algorithm GradMDM. GradMDM is a technique that adjusts the direction and the magnitude of the gradients to effectively find a small perturbation for each input, that will activate more computational units of dynamic models during inference. We evaluate GradMDM on multiple datasets and dynamic models, where it outperforms previous energy-oriented attack techniques, significantly increasing computation complexity while reducing the perceptibility of the perturbations.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 44 of its 62 references that are in the map. First posted on arXiv 1 Apr 2023.

Builds on

44 of 62 references.

References · 44

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  5. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  6. 2020End-to-End Object Detection with Transformerscited by 7,414
  7. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  8. 2017Mask R-CNNcited by 7,277
  9. 2016Densely Connected Convolutional Networkscited by 6,720
  10. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  11. 2013Intriguing properties of neural networkscited by 5,155
  12. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,859
32 more
  1. 2022A ConvNet for the 2020scited by 3,579
  2. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,241
  3. 2015SSD: Single Shot MultiBox Detectorcited by 3,130
  4. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  5. 2015Fast R-CNNcited by 2,933
  6. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,079
  7. 2015SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentationcited by 1,487
  8. 2016Adaptive Computation Time for Recurrent Neural Networkscited by 506
  9. 2017Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networkscited by 486
  10. 2021Dynamic Neural Networks: A Surveycited by 408
  11. 2021AdaViT: Adaptive Tokens for Efficient Vision Transformercited by 362
  12. 2017SkipNet: Learning Dynamic Routing in Convolutional Networkscited by 332
  13. 2017A Survey of Model Compression and Acceleration for Deep Neural Networkscited by 301
  14. 2019CondConv: Conditionally Parameterized Convolutions for Efficient Inferencecited by 273
  15. 2017BlockDrop: Dynamic Inference Paths in Residual Networkscited by 189
  16. 2015Conditional Computation in Neural Networks for faster modelscited by 167
  17. 2017Convolutional Networks with Adaptive Inference Graphscited by 148
  18. 2021Not All Images are Worth 16x16 Words: Dynamic Transformers for Efficient Image Recognitioncited by 144
  19. 2016Spatially Adaptive Computation Time for Residual Networkscited by 118
  20. 2017Keeping the Bad Guys Out: Protecting and Vaccinating Deep Learning with JPEG Compressioncited by 87
  21. 2021Dynamic Slimmable Networkcited by 83
  22. 2018Channel Gating Neural Networkscited by 80
  23. 2020Learning Dynamic Routing for Semantic Segmentationcited by 79
  24. 2019AdaBits: Neural Network Quantization with Adaptive Bit-Widthscited by 64
  25. 2020A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inferencecited by 61
  26. 2017Dynamic Deep Neural Networks: Optimizing Accuracy-Efficiency Trade-offs by Selective Executioncited by 59
  27. 2021Manifold Regularized Dynamic Network Pruningcited by 43
  28. 2019Any-Precision Deep Neural Networkscited by 40
  29. 2014Exponentially Increasing the Capacity-to-Computation Ratio for Conditional Computation in Deep Learningcited by 24
  30. 2020Fractional Skipping: Towards Finer-Grained Dynamic CNN Inferencecited by 23
  31. 20202D or not 2D? Adaptive 3D Convolution Selection for Efficient Video Recognitioncited by 22
  32. 2018IamNN: Iterative and Adaptive Mobile Neural Network for Efficient Image Classificationcited by 12

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 3; 2025: 2; 2026: 2.

8 papers.

  1. 2024Learning to Unlearn for Robust Machine Unlearningcited by 19
  2. 2024Robust and Transferable Backdoor Attacks Against Deep Image Compression With Selective Frequency Priorcited by 11
  3. 2025A Survey on Dynamic Neural Networks: from Computer Vision to Multi-modal Sensor Fusioncited by 5
  4. 2024Resilience of Entropy Model in Distributed Neural Networkscited by 1
  5. 2025Efficiency Robustness of Dynamic Deep Learning Systemscited by 1
  6. 2026Jellyfish: Zero-Shot Federated Unlearning Scheme with Knowledge Disentanglementcited by 1
  7. 2023Trustworthy Large Models in Vision: A Surveycited by 0
  8. 2026AESOP: Adversarial Execution-path Selection to Overload Deep Learning Pipelinescited by 0