mlsanityData to 7 Oct 2026

Towards Robust Stacked Capsule Autoencoder with Hybrid Adversarial Training

arXiv 2202.13755 ↗First posted 28 Feb 2022cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Capsule networks (CapsNets) are new neural networks that classify images based on the spatial relationships of features. By analyzing the pose of features and their relative positions, it is more capable to recognize images after affine transformation. The stacked capsule autoencoder (SCAE) is a state-of-the-art CapsNet, and achieved unsupervised classification of CapsNets for the first time. However, the security vulnerabilities and the robustness of the SCAE has rarely been explored. In this paper, we propose an evasion attack against SCAE, where the attacker can generate adversarial perturbations based on reducing the contribution of the object capsules in SCAE related to the original category of the image. The adversarial perturbations are then applied to the original images, and the perturbed images will be misclassified. Furthermore, we propose a defense method called Hybrid Adversarial Training (HAT) against such evasion attacks. HAT makes use of adversarial training and adversarial distillation to achieve better robustness and stability. We evaluate the defense method and the experimental results show that the refined SCAE model can achieve 82.14% classification accuracy under evasion attack. The source code is available at https://github.com/FrostbiteXSW/SCAE_Defense.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 43 of its 66 references that are in the map. First posted on arXiv 28 Feb 2022.

Builds on

43 of 66 references.

References · 43

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  4. 2013Intriguing properties of neural networkscited by 5,114
  5. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  6. 2016Adversarial examples in the physical worldcited by 1,762
  7. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  8. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
  9. 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
  10. 2015The Limitations of Deep Learning in Adversarial Settingscited by 784
  11. 2016Universal adversarial perturbationscited by 774
  12. 2017Dynamic Routing Between Capsulescited by 689
31 more
  1. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
  2. 2019Adversarial Training for Free!cited by 638
  3. 2017One pixel attack for fooling deep neural networkscited by 579
  4. 2017Countering Adversarial Images using Input Transformationscited by 517
  5. 2018Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networkscited by 500
  6. 2017Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Modelscited by 459
  7. 2018Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Surveycited by 459
  8. 2017Mitigating Adversarial Effects Through Randomizationcited by 382
  9. 2019Hidden Trigger Backdoor Attackscited by 359
  10. 2017Certifying Some Distributional Robustness with Principled Adversarial Trainingcited by 333
  11. 2016Adversarial Training Methods for Semi-Supervised Text Classificationcited by 280
  12. 2019Adversarial Attacks and Defenses in Images, Graphs and Text: A Reviewcited by 234
  13. 2019A backdoor attack against LSTM-based text classification systemscited by 214
  14. 2018Stochastic Activation Pruning for Robust Adversarial Defensecited by 171
  15. 2018Backdoor Embedding in Convolutional Neural Network Models via Invisible Perturbationcited by 131
  16. 2019Transferable Clean-Label Poisoning Attacks on Deep Neural Netscited by 129
  17. 2019Adversarially Robust Distillationcited by 124
  18. 2019Stacked Capsule Autoencoderscited by 82
  19. 2017Adversarial Transformation Networks: Learning to Generate Adversarial Examplescited by 60
  20. 2021Reliable Adversarial Distillation with Unreliable Teacherscited by 55
  21. 2017Houdini: Fooling Deep Structured Prediction Modelscited by 50
  22. 2019Gotta Catch 'Em All: Using Honeypots to Catch Adversarial Attacks on Neural Networkscited by 40
  23. 2017UPSET and ANGRI : Breaking High Performance Image Classifierscited by 19
  24. 2020A Sweet Rabbit Hole by DARCY: Using Honeypots to Detect Universal Trigger's Adversarial Attackscited by 14
  25. 2019On the Vulnerability of Capsule Networks to Adversarial Attackscited by 11
  26. 2018DARCCC: Detecting Adversaries by Reconstruction from Class Conditional Capsulescited by 9
  27. 2019CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networkscited by 8
  28. 2020Deflecting Adversarial Attackscited by 7
  29. 2018Increasing the adversarial robustness and explainability of capsule networks with γ-capsulescited by 5
  30. 2020Detecting Adversarial Patches with Class Conditional Reconstruction Networkscited by 2
  31. 2020An Evasion Attack against Stacked Capsule Autoencodercited by 1

Built on by

Not cited yet.