mlsanityData to 7 Oct 2026

Understanding Adversarial Examples Through Deep Neural Network's Response Surface and Uncertainty Regions

arXiv 2107.00003 ↗First posted 30 Jun 2021cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Deep neural network (DNN) is a popular model implemented in many systems to handle complex tasks such as image classification, object recognition, natural language processing etc. Consequently DNN structural vulnerabilities become part of the security vulnerabilities in those systems. In this paper we study the root cause of DNN adversarial examples. We examine the DNN response surface to understand its classification boundary. Our study reveals the structural problem of DNN classification boundary that leads to the adversarial examples. Existing attack algorithms can generate from a handful to a few hundred adversarial examples given one clean image. We show there are infinitely many adversarial images given one clean sample, all within a small neighborhood of the clean sample. We then define DNN uncertainty regions and show transferability of adversarial examples is not universal. We also argue that generalization error, the large sample theoretical guarantee established for DNN, cannot adequately capture the phenomenon of adversarial examples. We need new theory to measure DNN robustness.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 33 of its 72 references that are in the map. First posted on arXiv 30 Jun 2021.

Builds on

33 of 72 references.

References · 33

  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. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  5. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  6. 2016Adversarial examples in the physical worldcited by 1,762
  7. 2016Understanding deep learning requires rethinking generalizationcited by 1,657
  8. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  9. 2017Spectrally-normalized margin bounds for neural networkscited by 681
  10. 2017Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Modelscited by 459
  11. 2018Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Surveycited by 459
  12. 2016Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samplescited by 439
21 more
  1. 2017Wild Patterns: Ten Years After the Rise of Adversarial Machine Learningcited by 411
  2. 2018Adversarial Attacks on Neural Networks for Graph Datacited by 395
  3. 2018Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifierscited by 341
  4. 2017Adversarial Attacks on Neural Network Policiescited by 314
  5. 2017A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networkscited by 309
  6. 2015Norm-Based Capacity Control in Neural Networkscited by 299
  7. 2017Size-Independent Sample Complexity of Neural Networkscited by 289
  8. 2018Spatially Transformed Adversarial Examplescited by 195
  9. 2017Tactics of Adversarial Attack on Deep Reinforcement Learning Agentscited by 163
  10. 2017The Space of Transferable Adversarial Examplescited by 156
  11. 2017Robust Deep Reinforcement Learning with Adversarial Attackscited by 152
  12. 2016Crafting Adversarial Input Sequences for Recurrent Neural Networkscited by 102
  13. 2017Vulnerability of Deep Reinforcement Learning to Policy Induction Attackscited by 91
  14. 2019Transferring Robustness for Graph Neural Network Against Poisoning Attackscited by 58
  15. 2018Generalization Error in Deep Learningcited by 46
  16. 2016Generalization Error of Invariant Classifierscited by 42
  17. 2018On Tighter Generalization Bound for Deep Neural Networks: CNNs, ResNets, and Beyondcited by 24
  18. 2019Detecting Adversarial Examples through Nonlinear Dimensionality Reductioncited by 3
  19. 2019Adaptive Gradient for Adversarial Perturbations Generationcited by 2
  20. 2019Fast generalization error bound of deep learning without scale invariance of activation functionscited by 2
  21. 2020Dynamically Computing Adversarial Perturbations for Recurrent Neural Networkscited by 1

Built on by

Not cited yet.