mlsanityData to 7 Oct 2026

A Perturbation Resistant Transformation and Classification System for Deep Neural Networks

arXiv 2208.11839 ↗First posted 25 Aug 2022cs.CV cs.LG cs.NEcited by 0

Not on the map. Why?

Abstract (arXiv)

Deep convolutional neural networks accurately classify a diverse range of natural images, but may be easily deceived when designed, imperceptible perturbations are embedded in the images. In this paper, we design a multi-pronged training, input transformation, and image ensemble system that is attack agnostic and not easily estimated. Our system incorporates two novel features. The first is a transformation layer that computes feature level polynomial kernels from class-level training data samples and iteratively updates input image copies at inference time based on their feature kernel differences to create an ensemble of transformed inputs. The second is a classification system that incorporates the prediction of the undefended network with a hard vote on the ensemble of filtered images. Our evaluations on the CIFAR10 dataset show our system improves the robustness of an undefended network against a variety of bounded and unbounded white-box attacks under different distance metrics, while sacrificing little accuracy on clean images. Against adaptive full-knowledge attackers creating end-to-end attacks, our system successfully augments the existing robustness of adversarially trained networks, for which our methods are most effectively applied.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 30 of its 36 references that are in the map. First posted on arXiv 25 Aug 2022.

Builds on

30 of 36 references.

References · 30

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2013Intriguing properties of neural networkscited by 5,114
  6. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  7. 2016Adversarial examples in the physical worldcited by 1,762
  8. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  9. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  10. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  11. 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
  12. 2019Adversarial Training for Free!cited by 638
18 more
  1. 2017Countering Adversarial Images using Input Transformationscited by 517
  2. 2017Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networkscited by 484
  3. 2015A Neural Algorithm of Artistic Stylecited by 480
  4. 2017Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Modelscited by 459
  5. 2019On Evaluating Adversarial Robustnesscited by 442
  6. 2017Mitigating Adversarial Effects Through Randomizationcited by 382
  7. 2017PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examplescited by 243
  8. 2018MMA Training: Direct Input Space Margin Maximization through Adversarial Trainingcited by 171
  9. 2019Defense Against Adversarial Attacks Using Feature Scattering-based Adversarial Trainingcited by 95
  10. 2018Shield: Fast, Practical Defense and Vaccination for Deep Learning using JPEG Compressioncited by 74
  11. 2019Image Super-Resolution as a Defense Against Adversarial Attackscited by 64
  12. 2017APE-GAN: Adversarial Perturbation Elimination with GANcited by 57
  13. 2019Adversarial Defense by Restricting the Hidden Space of Deep Neural Networkscited by 56
  14. 2017MagNet and "Efficient Defenses Against Adversarial Attacks" are Not Robust to Adversarial Examplescited by 46
  15. 2017Adversarial Example Defenses: Ensembles of Weak Defenses are not Strongcited by 32
  16. 2017Robustness to Adversarial Examples through an Ensemble of Specialistscited by 21
  17. 2017Cascade Adversarial Machine Learning Regularized with a Unified Embeddingcited by 13
  18. 2020Stylized Adversarial Defensecited by 5

Built on by

Not cited yet.