Robustness Verification of Deep Neural Networks on High-speed Rail Operating Environment Recognition
CSTR:
Author:
Affiliation:

School of Software Engineering, Tongji University, Shanghai 201804, China

Clc Number:

TP389.1

  • Article
  • | |
  • Metrics
  • |
  • Reference [18]
  • |
  • Related
  • | | |
  • Comments
    Abstract:

    The implementation of DeepTRE was improved to adapt to large-scale dataset scenarios, which greatly reduces the space complexity of DeepTRE when retaining the excellent verification ability of DeepTRE. The improved DeepTRE was evaluated in the high-speed rail operating environment recognition scenarios and was compared with other mainstream verification tools, i.e.,DLV and SafeCV. The experimental results show that the memory usage of the improved DeepTRE tool is significantly lower than that of the original DeepTRE tool. Compared with other neural network verification tools, the improved DeepTRE tool has better verification effect on the premise of faster verification speed.

    Reference
    [1] RUSSELL S J. Artificial intelligence: a modern approach[M]. New York: Pearson Education, Inc., 2010.
    [2] MITCHELL T M. Machine learning[M]. New York: McGraw-Hill Co. Ltd., 1997.
    [3] SESHIA S A, SADIGH D, SHANKAR SASTRY S. Towards verified artificial intelligence[J]. Communications of the ACM, 2016, 65(7): 46.
    [4] DIETTERICH T G, HORVITZ E J. Rise of concerns about AI: reflections and directions[J]. Communications of the ACM, 2015, 58(10): 38.
    [5] SZEGEDY C, ZAREMBA W, SUTSKEVER I, et al. Intriguing properties of neural networks[J/OL]. [2022-02-01]. https://arxiv.org/abs/1312.6199.
    [6] RUAN W, HUANG X, KWIATKOWSKA M. Reachability analysis of deep neural networks with provable guarantees[C]//IJCAI. Stockholm: IJCAI, 2018: 2651-2659.
    [7] CLARKE E M, WING J M. Formal methods: state of the art and future directions[J]. ACM Computing Surveys (CSUR), 1996, 28(4): 626.
    [8] WING J M. A specifier’s introduction to formal methods[J]. Computer, 1990, 23(9): 8.
    [9] KATZ G, BARRETT C, DILL D L, et al. Reluplex: an efficient SMT solver for verifying deep neural networks[C]//International Conference on Computer Aided Verification. Heidelberg: Springer, 2017: 97-117.
    [10] LOMUSCIO A, MAGANTI L. An approach to reachability analysis for feed-forward ReLU neural networks[J/OL]. [2022-02-01]. https://arxiv.org/abs/1706.07351.
    [11] DUTTA S, JHA S, SANKARANARAYANAN S, et al. Output range analysis for deep feedforward neural networks[C]//NASA Formal Methods Symposium. Heidelberg:Springer, 2018: 121-138.
    [12] HUANG X, KWIATKOWSKA M, WANG S, et al. Safety verification of deep neural networks[C]//International Conference on Computer Aided Verification. Heidelberg: Springer, 2017: 3-29.
    [13] WICKER M, HUANG X, KWIATKOWSKA M. Feature-guided black-box safety testing of deep neural networks[C]//International Conference on Tools and Algorithms for the Construction and Analysis of Systems. Heidelberg: Springer, 2018: 408-426.
    [14] MOOSAVI-DEZFOOLI S M, FAWZI A, FROSSARD P. DeepFool: a simple and accurate method to fool deep neural networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE Computer Society, 2016: 2574-2582.
    [15] GEHR T, MIRMAN M, DRACHSLER-COHEN D, et al. AI2: safety and robustness certification of neural networks with abstract interpretation[C]//2018 IEEE Symposium on Security and Privacy (SP). San Francisco: IEEE, 2018: 3-18.
    [16] RUAN W, WU M, SUN Y, et al. Global robustness evaluation of deep neural networks with provable guarantees for the hamming distance[C]//IJCAI. Macao: IJCAI,2019: 5944-5952.
    [17] CARLINI N, WAGNER D. Towards evaluating the robustness of neural networks[C]//2017 IEEE Symposium on Security and Privacy (sp). San Jose: IEEE Computer Society, 2017: 39-57.
    [18] HOU X, AN J, ZHANG M, et al. High-speed rail operating environment recognition based on neural network and adversarial training[C]//2019 IEEE 31st International Conference on Tools with Artificial Intelligence (ICTAI). Portland: IEEE Computer Society, 2019: 840-847.
    Related
    Cited by
    Comments
    Comments
    分享到微博
    Submit
Get Citation

GAO Zhen, SU Yu, HOU Xiaoxue, FANG Pei, ZHANG Miaomiao. Robustness Verification of Deep Neural Networks on High-speed Rail Operating Environment Recognition[J].同济大学学报(自然科学版),2022,50(10):1405~1413

Copy
Share
Article Metrics
  • Abstract:1532
  • PDF: 436
  • HTML: 87
  • Cited by: 0
History
  • Received:May 10,2022
  • Online: November 03,2022
Article QR Code