mlsanityData to 7 Oct 2026

Knowledge Enhanced Machine Learning Pipeline against Diverse Adversarial Attacks

arXiv 2106.06235 ↗First posted 11 Jun 2021cs.LGcited by 24

Not on the map. Why?

Abstract (arXiv)

Despite the great successes achieved by deep neural networks (DNNs), recent studies show that they are vulnerable against adversarial examples, which aim to mislead DNNs by adding small adversarial perturbations. Several defenses have been proposed against such attacks, while many of them have been adaptively attacked. In this work, we aim to enhance the ML robustness from a different perspective by leveraging domain knowledge: We propose a Knowledge Enhanced Machine Learning Pipeline (KEMLP) to integrate domain knowledge (i.e., logic relationships among different predictions) into a probabilistic graphical model via first-order logic rules. In particular, we develop KEMLP by integrating a diverse set of weak auxiliary models based on their logical relationships to the main DNN model that performs the target task. Theoretically, we provide convergence results and prove that, under mild conditions, the prediction of KEMLP is more robust than that of the main DNN model. Empirically, we take road sign recognition as an example and leverage the relationships between road signs and their shapes and contents as domain knowledge. We show that compared with adversarial training and other baselines, KEMLP achieves higher robustness against physical attacks, Lp bounded attacks, unforeseen attacks, and natural corruptions under both whitebox and blackbox settings, while still maintaining high clean accuracy.

Key facts. Cited by 24 papers in the map (data through 7 Oct 2026). Builds on 43 of its 72 references that are in the map. First posted on arXiv 11 Jun 2021.

Builds on

43 of 72 references.

References · 43

  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. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  5. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  6. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  7. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  8. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 950
  9. 2018Robustness May Be at Odds with Accuracycited by 873
  10. 2019Adversarial Training for Free!cited by 638
  11. 2017Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methodscited by 542
  12. 2017Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networkscited by 484
31 more
  1. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  2. 2017Snorkel: Rapid Training Data Creation with Weak Supervisioncited by 370
  3. 2018Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Modelscited by 362
  4. 2019Provably Robust Deep Learning via Adversarially Trained Smoothed Classifierscited by 328
  5. 2018On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Modelscited by 322
  6. 2018Generating Adversarial Examples with Adversarial Networkscited by 256
  7. 2016Data Programming: Creating Large Training Sets, Quicklycited by 246
  8. 2017Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradientscited by 236
  9. 2018Adversarial Risk and the Dangers of Evaluating Against Weak Attackscited by 221
  10. 2019Improving Adversarial Robustness via Promoting Ensemble Diversitycited by 213
  11. 2018Spatially Transformed Adversarial Examplescited by 195
  12. 2018Certified Adversarial Robustness with Additive Noisecited by 178
  13. 2017Towards Robust Neural Networks via Random Self-ensemblecited by 158
  14. 2020Understanding and Mitigating the Tradeoff Between Robustness and Accuracycited by 148
  15. 2020Randomized Smoothing of All Shapes and Sizescited by 146
  16. 2020MACER: Attack-free and Scalable Robust Training via Maximizing Certified Radiuscited by 114
  17. 2018Towards the first adversarially robust neural network model on MNISTcited by 104
  18. 2018Curriculum Adversarial Trainingcited by 89
  19. 2019Wasserstein Adversarial Examples via Projected Sinkhorn Iterationscited by 89
  20. 2013Multi-digit Number Recognition from Street View Imagery using Deep Convolutional Neural Networkscited by 84
  21. 2019Unrestricted Adversarial Examples via Semantic Manipulationcited by 84
  22. 2020Precise Tradeoffs in Adversarial Training for Linear Regressioncited by 76
  23. 2020DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensemblescited by 73
  24. 2019Improving Adversarial Robustness of Ensembles with Diversity Trainingcited by 69
  25. 2019Defending Against Physically Realizable Attacks on Image Classificationcited by 66
  26. 2019Joint Inference of Reward Machines and Policies for Reinforcement Learningcited by 56
  27. 2020Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Frameworkcited by 45
  28. 2018Characterizing Adversarial Examples Based on Spatial Consistency Information for Semantic Segmentationcited by 40
  29. 2017Ensemble Methods as a Defense to Adversarial Perturbations Against Deep Neural Networkscited by 31
  30. 2020Hidden Cost of Randomized Smoothingcited by 5
  31. 2014Joint Inference of Multiple Label Types in Large Networkscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 8; 2023: 5; 2024: 4; 2025: 6; 2026: 1.

24 papers.

Sort:

By area:

  1. 2025ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoningcited by 144
  2. 2022Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbationscited by 58
  3. 2024R2-Guard: Robust Reasoning Enabled LLM Guardrail via Knowledge-Enhanced Logical Reasoningcited by 41
  4. 2022Recognizing Object by Components with Human Prior Knowledge Enhances Adversarial Robustness of Deep Neural Networkscited by 18
  5. 2023Mitigating Accuracy-Robustness Trade-off via Balanced Multi-Teacher Adversarial Distillationcited by 13
  6. 2023Towards Building More Robust Models with Frequency Biascited by 12
  7. 2022Visual Adversarial Attacks and Defenses in the Physical World: A Surveycited by 11
  8. 2022CARE: Certifiably Robust Learning with Reasoning via Variational Inferencecited by 10
  9. 2024COLEP: Certifiably Robust Learning-Reasoning Conformal Prediction via Probabilistic Circuitscited by 9
  10. 2022Game Theoretic Mixed Experts for Combinational Adversarial Machine Learningcited by 6
  11. 2025Neural Probabilistic Circuits: Enabling Compositional and Interpretable Predictions through Logical Reasoningcited by 6
  12. 2022Informed Pre-Training on Prior Knowledgecited by 5
12 more
  1. 2023Revisiting the Trade-off between Accuracy and Robustness via Weight Distribution of Filterscited by 4
  2. 2024Improving deep learning with prior knowledge and cognitive models: A survey on enhancing explainability, adversarial robustness and zero-shot learningcited by 4
  3. 2024KnowGraph: Knowledge-Enabled Anomaly Detection via Logical Reasoning on Graph Datacited by 4
  4. 2022Verifiable Obstacle Detectioncited by 3
  5. 2023Distributional Modeling for Location-Aware Adversarial Patchescited by 3
  6. 2022Knowledge is Power: Understanding Causality Makes Legal judgment Prediction Models More Generalizable and Robustcited by 2
  7. 2025Factor Graph-based Interpretable Neural Networkscited by 2
  8. 2025Understanding and Improving Adversarial Robustness of Neural Probabilistic Circuitscited by 1
  9. 2026Knowledge-Guided Adversarial Training for Infrared Object Detection via Thermal Radiation Modelingcited by 1
  10. 2023Epistemic Graph: A Plug-And-Play Module For Hybrid Representation Learningcited by 0
  11. 2025DARD: Dice Adversarial Robustness Distillation against Adversarial Attackscited by 0
  12. 2025VISAT: Benchmarking Adversarial and Distribution Shift Robustness in Traffic Sign Recognition with Visual Attributescited by 0