mlsanityData to 7 Oct 2026

Tighter Abstract Queries in Neural Network Verification

arXiv 2210.12871 ↗First posted 23 Oct 2022cs.LG cs.LO cs.NEcited by 8

Not on the map. Why?

Abstract (arXiv)

Neural networks have become critical components of reactive systems in various domains within computer science. Despite their excellent performance, using neural networks entails numerous risks that stem from our lack of ability to understand and reason about their behavior. Due to these risks, various formal methods have been proposed for verifying neural networks; but unfortunately, these typically struggle with scalability barriers. Recent attempts have demonstrated that abstraction-refinement approaches could play a significant role in mitigating these limitations; but these approaches can often produce networks that are so abstract, that they become unsuitable for verification. To deal with this issue, we present CEGARETTE, a novel verification mechanism where both the system and the property are abstracted and refined simultaneously. We observe that this approach allows us to produce abstract networks which are both small and sufficiently accurate, allowing for quick verification times while avoiding a large number of refinement steps. For evaluation purposes, we implemented CEGARETTE as an extension to the recently proposed CEGAR-NN framework. Our results are very promising, and demonstrate a significant improvement in performance over multiple benchmarks.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 53 of its 100 references that are in the map. First posted on arXiv 23 Oct 2022.

Builds on

53 of 100 references.

References · 53

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  4. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,838
  5. 2016End to End Learning for Self-Driving Carscited by 1,000
  6. 2017Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networkscited by 739
  7. 2017Provable defenses against adversarial examples via the convex outer adversarial polytopecited by 564
  8. 2018Efficient Neural Network Robustness Certification with General Activation Functionscited by 439
  9. 2018On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Modelscited by 322
  10. 2017DeepXplore: Automated Whitebox Testing of Deep Learning Systemscited by 307
  11. 2016Safety Verification of Deep Neural Networkscited by 262
  12. 2017DeepTest: Automated Testing of Deep-Neural-Network-driven Autonomous Carscited by 249
38 more
  1. 2017Formal Verification of Piece-Wise Linear Feed-Forward Neural Networkscited by 241
  2. 2019Adversarial Attacks and Defenses in Images, Graphs and Text: A Reviewcited by 234
  3. 2019Algorithms for Verifying Deep Neural Networkscited by 221
  4. 2018Formal Security Analysis of Neural Networks using Symbolic Intervalscited by 185
  5. 2018A Study on Overfitting in Deep Reinforcement Learningcited by 170
  6. 2019Branch and Bound for Piecewise Linear Neural Network Verificationcited by 138
  7. 2017An approach to reachability analysis for feed-forward ReLU neural networkscited by 126
  8. 2018A Dual Approach to Scalable Verification of Deep Networkscited by 125
  9. 2021Video Summarization Using Deep Neural Networks: A Surveycited by 85
  10. 2017Output Reachable Set Estimation and Verification for Multi-Layer Neural Networkscited by 79
  11. 2017Verifying Properties of Binarized Deep Neural Networkscited by 77
  12. 2021PRIMA: General and Precise Neural Network Certification via Scalable Convex Hull Approximationscited by 77
  13. 2021The Second International Verification of Neural Networks Competition (VNN-COMP 2021): Summary and Resultscited by 71
  14. 2019An Abstraction-Based Framework for Neural Network Verificationcited by 68
  15. 2020An SMT-Based Approach for Verifying Binarized Neural Networkscited by 47
  16. 2021Efficient-CapsNet: Capsule Network with Self-Attention Routingcited by 41
  17. 2019Optimization and Abstraction: A Synergistic Approach for Analyzing Neural Network Robustnesscited by 40
  18. 2019Survey on Deep Neural Networks in Speech and Vision Systemscited by 37
  19. 2020Parallelization Techniques for Verifying Neural Networkscited by 36
  20. 2021Neural Network Robustness as a Verification Property: A Principled Case Studycited by 35
  21. 2020Abstraction based Output Range Analysis for Neural Networkscited by 34
  22. 2021Towards Scalable Verification of Deep Reinforcement Learningcited by 33
  23. 2022Verifying Learning-Based Robotic Navigation Systemscited by 33
  24. 2020DeepAbstract: Neural Network Abstraction for Accelerating Verificationcited by 31
  25. 2017Output Range Analysis for Deep Neural Networkscited by 30
  26. 2020Improving Neural Network Verification through Spurious Region Guided Refinementcited by 30
  27. 2020Verifying Recurrent Neural Networks using Invariant Inferencecited by 25
  28. 2020Global Optimization of Objective Functions Represented by ReLU Networkscited by 24
  29. 2022An Abstraction-Refinement Approach to Verifying Convolutional Neural Networkscited by 24
  30. 2022Efficient Neural Network Analysis with Sum-of-Infeasibilitiescited by 24
  31. 2022Neural Network Verification with Proof Productioncited by 22
  32. 2021Pruning and Slicing Neural Networks using Formal Verificationcited by 20
  33. 2022Verification-Aided Deep Ensemble Selectioncited by 18
  34. 2022Neural Network Verification using Residual Reasoningcited by 18
  35. 2021Minimal Multi-Layer Modifications of Deep Neural Networkscited by 14
  36. 2018Toward Scalable Verification for Safety-Critical Deep Networkscited by 12
  37. 2020Abstract Neural Networkscited by 11
  38. 2021DeepCert: Verification of Contextually Relevant Robustness for Neural Network Image Classifierscited by 11

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 2; 2025: 3; 2026: 2.

8 papers.

  1. 2025Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanationscited by 13
  2. 2023Verifying Generalization in Deep Learningcited by 11
  3. 2025Abstraction-Based Proof Production in Formal Verification of Neural Networkscited by 5
  4. 2024Towards Formal Verification of Deep Neural Networks for Object Detectioncited by 4
  5. 2024Safe and Reliable Training of Learning-Based Aerospace Controllerscited by 4
  6. 2025PICID: Proof-Driven Clause Learning in Neural Network Verificationcited by 3
  7. 2026Talking with Verifiers: Automatic Specification Generation for Neural Network Verificationcited by 0
  8. 2026Mining Verdict Boundaries for Neural Network Verificationcited by 0