Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
102 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

MISLEAD: Manipulating Importance of Selected features for Learning Epsilon in Evasion Attack Deception (2404.15656v2)

Published 24 Apr 2024 in cs.LG and cs.CR

Abstract: Emerging vulnerabilities in ML models due to adversarial attacks raise concerns about their reliability. Specifically, evasion attacks manipulate models by introducing precise perturbations to input data, causing erroneous predictions. To address this, we propose a methodology combining SHapley Additive exPlanations (SHAP) for feature importance analysis with an innovative Optimal Epsilon technique for conducting evasion attacks. Our approach begins with SHAP-based analysis to understand model vulnerabilities, crucial for devising targeted evasion strategies. The Optimal Epsilon technique, employing a Binary Search algorithm, efficiently determines the minimum epsilon needed for successful evasion. Evaluation across diverse machine learning architectures demonstrates the technique's precision in generating adversarial samples, underscoring its efficacy in manipulating model outcomes. This study emphasizes the critical importance of continuous assessment and monitoring to identify and mitigate potential security risks in machine learning systems.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (45)
  1. Intriguing properties of neural networks. 2nd International Conference on Learning Representations (ICLR), 2014.
  2. Explaining and harnessing adversarial examples. 3rd International Conference on Learning Representations, (ICLR), 2015. URL http://arxiv.org/abs/1412.6572.
  3. Towards evaluating the robustness of neural networks. IEEE Symposium on Security and Privacy (SP), pages 39–57, 2017. doi:10.1109/SP.2017.49.
  4. A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems 30, pages 4765–4774, 2017.
  5. Wild patterns: Ten years after the rise of adversarial machine learning. Pattern Recognition, 84:317–331, December 2018. ISSN 0031-3203.
  6. From local explanations to global understanding with explainable ai for trees. Nature machine intelligence, page 56–67, 05 2019.
  7. Practical black-box attacks against machine learning. Proceedings of the 2017 ACM on Asia Conference on Computer and Communications Security, page 506–519, 2017.
  8. Enhancing the transferability of adversarial attacks through variance tuning. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 1924–1933, June 2021.
  9. Adversarial attacks on neural network policies. arXiv, 2017.
  10. Training verified learners with learned verifiers. ArXiv, abs/1805.10265, 2018.
  11. Synthesizing robust adversarial examples. Proceedings of the 35th International Conference on Machine Learning, 80:284–293, 10–15 Jul 2018.
  12. Potential adversarial samples for white-box attacks. arXiv preprint arXiv:1912.06409, 2019.
  13. Deep models under the gan: Information leakage from collaborative deep learning. Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security, page 603–618, 2017.
  14. Deep neural rejection against adversarial examples. EURASIP Journal on Information Security, 2020.
  15. From explanations to feature selection: assessing shap values as feature selection mechanism. 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), pages 340–347, 2020. doi:10.1109/SIBGRAPI51738.2020.00053.
  16. Feature relevance evaluation using grad-cam, lime and shap for deep learning sar data classification. 23rd International Radar Symposium (IRS), pages 457–462, 2022.
  17. SHAP-based explanation methods: A review for NLP interpretability. Proceedings of the 29th International Conference on Computational Linguistics, pages 4593–4603, October 2022.
  18. From local explanations to global understanding with explainable ai for trees. Nature Machine Intelligence, 2(1):2522–5839, 2020.
  19. Feature selection in machine learning: A new perspective. Neurocomputing, 300:70–79, 2018. ISSN 0925-2312.
  20. Learning to explain: An information-theoretic perspective on model interpretation. In Proceedings of the 37th International Conference on Machine Learning (ICML), 80, 2018.
  21. Towards better understanding of gradient-based attribution methods for deep neural networks. International Conference on Learning Representations (ICLR), 2018.
  22. Binary search based boundary elimination selection in many-objective evolutionary optimization. Applied Soft Computing, 60:689–705, 2017.
  23. Bo Han and Yongquan Lu. Research on optimization and parallelization of optimal binary search tree using dynamic programming. Proceedings of the 2nd International Conference on Electronic and Mechanical Engineering and Information Technology (EMEIT 2012), pages 229–233, 2012/09.
  24. Safety-critical computer vision: an empirical survey of adversarial evasion attacks and defenses on computer vision systems. Artificial Intelligence Review, pages 1–35, 06 2023.
  25. Towards deep learning models resistant to adversarial attacks. 6th International Conference on Learning Representations, ICLR Conference Track Proceedings, 06 2017a.
  26. Deepfool: A simple and accurate method to fool deep neural networks. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2574–2582, 2016. doi:10.1109/CVPR.2016.282.
  27. Permuteattack: Counterfactual explanation of machine learning credit scorecards. arXiv:2008.10138, 2020.
  28. Robust classification of financial risk. arXiv:1811.11079, 2018.
  29. Adversarial attacks for tabular data: Application to fraud detection and imbalanced data. arXiv:2101.08030, 2021.
  30. Feature importance guided attack: A model agnostic adversarial attack. arXiv preprint arXiv:2106.14815, 2023.
  31. Interpreting black-box models: A review on explainable artificial intelligence. Cognitive Computation, 16(1):45–74, 2024. doi:10.1007/s12559-023-10179-8. URL https://doi.org/10.1007/s12559-023-10179-8.
  32. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12:2825–2830, 2011.
  33. R. A. Fisher. Iris. UCI Machine Learning Repository, 1988. DOI: https://doi.org/10.24432/C56C76.
  34. The Iris Data Set: In Search of the Source of Virginica. Significance, 18(6):26–29, 11 2021. URL https://doi.org/10.1111/1740-9713.01589.
  35. Bank Marketing. UCI Machine Learning Repository, 2012. https://doi.org/10.24432/C5K306.
  36. A data-driven approach to predict the success of bank telemarketing. Decision Support Systems, 62:22–31, 2014.
  37. A training algorithm for optimal margin classifiers. Proceedings of the Fifth Annual Workshop on Computational Learning Theory, page 144–152, 1992.
  38. Xgboost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, page 785–794, 2016.
  39. Adversarial robustness toolbox v1.2.0. CoRR, 1807.01069, 2018. URL https://arxiv.org/pdf/1807.01069.
  40. secml: A python library for secure and explainable machine learning. arXiv preprint arXiv:1912.10013, 2019.
  41. Grad-cam: Visual explanations from deep networks via gradient-based localization. In 2017 IEEE International Conference on Computer Vision (ICCV), pages 618–626, 2017. doi:10.1109/ICCV.2017.74.
  42. Learning important features through propagating activation differences. In Proceedings of the 34th International Conference on Machine Learning - Volume 70, ICML’17, page 3145–3153. JMLR.org, 2017.
  43. Layer-Wise Relevance Propagation: An Overview, pages 193–209. Springer International Publishing, Cham, 2019.
  44. Towards deep learning models resistant to adversarial attacks. arXiv preprint arXiv:1706.06083, 2017b.
  45. Adversarially robust distillation. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 3996–4003, 2020.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (4)
  1. Vidit Khazanchi (2 papers)
  2. Pavan Kulkarni (6 papers)
  3. Yuvaraj Govindarajulu (8 papers)
  4. Manojkumar Parmar (15 papers)