mlsanityData to 7 Oct 2026

An Adversarial Robustness Perspective on the Topology of Neural Networks

arXiv 2211.02675 ↗First posted 4 Nov 2022cs.LG cs.CR stat.MLcited by 3

Not on the map. Why?

Abstract (arXiv)

In this paper, we investigate the impact of neural networks (NNs) topology on adversarial robustness. Specifically, we study the graph produced when an input traverses all the layers of a NN, and show that such graphs are different for clean and adversarial inputs. We find that graphs from clean inputs are more centralized around highway edges, whereas those from adversaries are more diffuse, leveraging under-optimized edges. Through experiments on a variety of datasets and architectures, we show that these under-optimized edges are a source of adversarial vulnerability and that they can be used to detect adversarial inputs.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 37 of its 59 references that are in the map. First posted on arXiv 4 Nov 2022.

Builds on

37 of 59 references.

References · 37

  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. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  5. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  6. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,250
  7. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,472
  8. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  9. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  10. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 950
  11. 2016Adversarial Machine Learning at Scalecited by 934
  12. 2016Practical Black-Box Attacks against Machine Learningcited by 893
25 more
  1. 2018Robustness May Be at Odds with Accuracycited by 873
  2. 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
  3. 2016Universal adversarial perturbationscited by 774
  4. 2015Cyclical Learning Rates for Training Neural Networkscited by 627
  5. 2017Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methodscited by 542
  6. 2020Overfitting in adversarially robust deep learningcited by 504
  7. 2017Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Modelscited by 459
  8. 2018Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionalitycited by 280
  9. 2019Deconstructing Lottery Tickets: Zeros, Signs, and the Supermaskcited by 206
  10. 2017On the (Statistical) Detection of Adversarial Examplescited by 180
  11. 2017An introduction to Topological Data Analysis: fundamental and practical aspects for data scientistscited by 171
  12. 2018Disentangling Adversarial Robustness and Generalizationcited by 171
  13. 2017The Space of Transferable Adversarial Examplescited by 156
  14. 2016Robust Large Margin Deep Neural Networkscited by 117
  15. 2018Improving DNN Robustness to Adversarial Attacks using Jacobian Regularizationcited by 88
  16. 2020Do Wider Neural Networks Really Help Adversarial Robustness?cited by 79
  17. 2017Sliced Wasserstein Kernel for Persistence Diagramscited by 77
  18. 2019Detecting Adversarial Samples Using Influence Functions and Nearest Neighborscited by 72
  19. 2018Sparse DNNs with Improved Adversarial Robustnesscited by 71
  20. 2021Excess Capacity and Backdoor Poisoningcited by 24
  21. 2019Adversarial Defense Through Network Profiling Based Path Extractioncited by 22
  22. 2019Interpreting Adversarial Examples by Activation Promotion and Suppressioncited by 17
  23. 2020Achieving Adversarial Robustness via Sparsitycited by 14
  24. 2017Adversary Detection in Neural Networks via Persistent Homologycited by 12
  25. 2019Characterizing the Shape of Activation Space in Deep Neural Networkscited by 3

Built on by

Cited by year

Citing papers per year: 2024: 2; 2025: 1.

3 papers.

  1. 2024When Witnesses Defend: A Witness Graph Topological Layer for Adversarial Graph Learningcited by 2
  2. 2024Be Persistent: Towards a Unified Solution for Mitigating Shortcuts in Deep Learningcited by 1
  3. 2025TopoReformer: Mitigating Adversarial Attacks Using Topological Purification in OCR Modelscited by 0