Verification for Neural Network Based on Error Divide and Conquer
Author:
Affiliation:

Clc Number:

Fund Project:

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    With the rapid development of neural network technology, neural networks have been widely applied in safety-critical fields such as autonomous driving, intelligent manufacturing, and medical diagnosis. Thus, it is crucial to ensure the trustworthiness of neural networks. However, due to the vulnerability of neural networks, slight perturbation often leads to wrong results. Therefore, it is vital to use formal verification methods to ensure the safety and trustworthiness of neural networks. Current verification methods for neural networks are mainly concerned with the accuracy of the analysis, while apt to ignore operational efficiency. When verifying the safety properties of complex networks, the large-scale state space may lead to problems such as infeasibility or unsolvability. To reduce the state space of neural networks and improve the verification efficiency, this study presents a formal verification method for neural networks based on divide and conquer considering over-approximation errors. The method uses the reachability analysis technique to calculate the upper and lower bounds of nonlinear nodes and uses an improved symbolic linear relaxation method to reduce over-approximation errors during the boundary calculation of nonlinear nodes. The constraints of nodes are refined by calculating the direct and indirect effects of their over-approximation errors. Thereby, the original verification problem is split into a set of sub-problems whose mixed integer linear programming (MILP) formulation has a smaller number of constraints. The method is implemented as a tool named NNVerifier, whose properties are verified and evaluated through experiments on four ReLU-based fully-connected benchmark networks trained on three classic datasets. The experimental results show that the verification efficiency of the NNVerifier is 37.18% higher than that of the existing complete verification methods.

    Reference
    Related
    Cited by
Get Citation

董彦松,刘月浩,董旭乾,赵亮,田聪,于斌,段振华.基于误差分治的神经网络验证.软件学报,2024,35(5):2307-2324

Copy
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:November 10,2022
  • Revised:January 20,2023
  • Adopted:
  • Online: September 27,2023
  • Published: May 06,2024
You are the firstVisitors
Copyright: Institute of Software, Chinese Academy of Sciences Beijing ICP No. 05046678-4
Address:4# South Fourth Street, Zhong Guan Cun, Beijing 100190,Postal Code:100190
Phone:010-62562563 Fax:010-62562533 Email:jos@iscas.ac.cn
Technical Support:Beijing Qinyun Technology Development Co., Ltd.

Beijing Public Network Security No. 11040202500063