mlsanityData to 7 Oct 2026

Increasing the Confidence of Deep Neural Networks by Coverage Analysis

arXiv 2101.12100 ↗First posted 28 Jan 2021cs.LG cs.AIcited by 9

Not on the map. Why?

Abstract (arXiv)

The great performance of machine learning algorithms and deep neural networks in several perception and control tasks is pushing the industry to adopt such technologies in safety-critical applications, as autonomous robots and self-driving vehicles. At present, however, several issues need to be solved to make deep learning methods more trustworthy, predictable, safe, and secure against adversarial attacks. Although several methods have been proposed to improve the trustworthiness of deep neural networks, most of them are tailored for specific classes of adversarial examples, hence failing to detect other corner cases or unsafe inputs that heavily deviate from the training samples. This paper presents a lightweight monitoring architecture based on coverage paradigms to enhance the model robustness against different unsafe inputs. In particular, four coverage analysis methods are proposed and tested in the architecture for evaluating multiple detection logics. Experimental results show that the proposed approach is effective in detecting both powerful adversarial examples and out-of-distribution inputs, introducing limited extra-execution time and memory requirements.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 43 of its 58 references that are in the map. First posted on arXiv 28 Jan 2021.

Builds on

43 of 58 references.

References · 43

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  5. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  7. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  8. 2013Intriguing properties of neural networkscited by 5,114
  9. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  10. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  11. 2017Billion-scale similarity search with GPUscited by 2,477
  12. 2016Adversarial examples in the physical worldcited by 1,762
31 more
  1. 2016Adversarial Machine Learning at Scalecited by 934
  2. 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
  3. 2015The Limitations of Deep Learning in Adversarial Settingscited by 784
  4. 2018Evidential Deep Learning to Quantify Classification Uncertaintycited by 728
  5. 2017Evasion Attacks against Machine Learning at Test Timecited by 686
  6. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
  7. 2017Provable defenses against adversarial examples via the convex outer adversarial polytopecited by 564
  8. 2017Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methodscited by 542
  9. 2017Adversarial Patchcited by 514
  10. 2015An Introduction to Convolutional Neural Networkscited by 510
  11. 2017Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networkscited by 484
  12. 2018Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Modelscited by 362
  13. 2017Certifying Some Distributional Robustness with Principled Adversarial Trainingcited by 333
  14. 2017DeepXplore: Automated Whitebox Testing of Deep Learning Systemscited by 307
  15. 2014Caffe: Convolutional Architecture for Fast Feature Embeddingcited by 287
  16. 2017MagNet: a Two-Pronged Defense against Adversarial Examplescited by 268
  17. 2018Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learningcited by 203
  18. 2017Formal Guarantees on the Robustness of a Classifier against Adversarial Manipulationcited by 182
  19. 2018Evaluating the Robustness of Neural Networks: An Extreme Value Theory Approachcited by 181
  20. 2018Are adversarial examples inevitable?cited by 113
  21. 2019Adversarial Defense by Restricting the Hidden Space of Deep Neural Networkscited by 56
  22. 2018Adversarial Sample Detection for Deep Neural Network through Model Mutation Testingcited by 48
  23. 2017MagNet and "Efficient Defenses Against Adversarial Attacks" are Not Robust to Adversarial Examplescited by 46
  24. 2018Safely Entering the Deep: A Review of Verification and Validation for Machine Learning and a Challenge Elicitation in the Automotive Industrycited by 38
  25. 2018Testing Deep Neural Networkscited by 36
  26. 2019Deep Neural Rejection against Adversarial Examplescited by 31
  27. 2021Detecting Adversarial Examples by Input Transformations, Defense Perturbations, and Votingcited by 16
  28. 2018MULDEF: Multi-model-based Defense Against Adversarial Examples for Neural Networkscited by 8
  29. 2018How the Softmax Output is Misleading for Evaluating the Strength of Adversarial Examplescited by 4
  30. 2019AdvKnn: Adversarial Attacks On K-Nearest Neighbor Classifiers With Approximate Gradientscited by 3
  31. 2018Controlling Over-generalization and its Effect on Adversarial Examples Generation and Detectioncited by 2

Built on by

Cited by year

Citing papers per year: 2022: 4; 2023: 3; 2024: 1; 2025: 1.

9 papers.

  1. 2022On the Real-World Adversarial Robustness of Real-Time Semantic Segmentation Models for Autonomous Drivingcited by 30
  2. 2022Defending From Physically-Realizable Adversarial Attacks Through Internal Over-Activation Analysiscited by 10
  3. 2022On the Minimal Adversarial Perturbation for Deep Neural Networks with Provable Estimation Errorcited by 6
  4. 2022Robust-by-Design Classification via Unitary-Gradient Neural Networkscited by 4
  5. 2025Multi-Layer Confidence Scoring for Detection of Out-of-Distribution Samples, Adversarial Attacks, and In-Distribution Misclassificationscited by 2
  6. 2023Detection of Adversarial Physical Attacks in Time-Series Image Datacited by 1
  7. 2023DREAM: Debugging and Repairing AutoML Pipelinescited by 1
  8. 2023Ensembling Uncertainty Measures to Improve Safety of Black-Box Classifierscited by 0
  9. 2024DeepCover: Advancing RNN Test Coverage and Online Error Prediction using State Machine Extractioncited by 0