mlsanityData to 7 Oct 2026

Graph-based methods coupled with specific distributional distances for adversarial attack detection

arXiv 2306.00042 ↗First posted 31 May 2023cs.LG cs.CR cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Artificial neural networks are prone to being fooled by carefully perturbed inputs which cause an egregious misclassification. These textitadversarial attacks have been the focus of extensive research. Likewise, there has been an abundance of research in ways to detect and defend against them. We introduce a novel approach of detection and interpretation of adversarial attacks from a graph perspective. For an input image, we compute an associated sparse graph using the layer-wise relevance propagation algorithm citebach15. Specifically, we only keep edges of the neural network with the highest relevance values. Three quantities are then computed from the graph which are then compared against those computed from the training set. The result of the comparison is a classification of the image as benign or adversarial. To make the comparison, two classification methods are introduced: 1) an explicit formula based on Wasserstein distance applied to the degree of node and 2) a logistic regression. Both classification methods produce strong results which lead us to believe that a graph-based interpretation of adversarial attacks is valuable.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 25 of its 43 references that are in the map. First posted on arXiv 31 May 2023.

Builds on

25 of 43 references.

References · 25

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  4. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  5. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,472
  6. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  7. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  8. 2014Striving for Simplicity: The All Convolutional Netcited by 983
  9. 2016Practical Black-Box Attacks against Machine Learningcited by 893
  10. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 682
  11. 2018Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionalitycited by 280
  12. 2017Detecting Adversarial Samples from Artifactscited by 244
13 more
  1. 2019Adversarial Attacks and Defenses in Images, Graphs and Text: A Reviewcited by 234
  2. 2018Adversarial Robustness Toolbox v1.0.0cited by 214
  3. 2020Topology of deep neural networkscited by 111
  4. 2019Wasserstein Adversarial Examples via Projected Sinkhorn Iterationscited by 89
  5. 2019The Odds are Odd: A Statistical Test for Detecting Adversarial Examplescited by 75
  6. 2016A topological insight into restricted Boltzmann machinescited by 32
  7. 2021SpectralDefense: Detecting Adversarial Attacks on CNNs in the Fourier Domaincited by 30
  8. 2020Topological Insights into Sparse Neural Networkscited by 21
  9. 2021SPADE: A Spectral Method for Black-Box Adversarial Robustness Evaluationcited by 16
  10. 2019Emergence of Network Motifs in Deep Neural Networkscited by 9
  11. 2020Representation Learning via Adversarially-Contrastive Optimal Transportcited by 4
  12. 2020Attack Agnostic Detection of Adversarial Examples via Random Subspace Analysiscited by 2
  13. 2022Deep Neural Networks as Complex Networkscited by 2

Built on by

Not cited yet.