Symbolic Execution for Importance Analysis and Adversarial Generation in Neural Networks

@article{Gopinath2019SymbolicEF,
  title={Symbolic Execution for Importance Analysis and Adversarial Generation in Neural Networks},
  author={Divya Gopinath and Mengshi Zhang and Kaiyuan Wang and Ismet Burak Kadron and Corina S. Pasareanu and Sarfraz Khurshid},
  journal={2019 IEEE 30th International Symposium on Software Reliability Engineering (ISSRE)},
  year={2019},
  pages={313-322}
}
  • Divya Gopinath, Mengshi Zhang, +3 authors Sarfraz Khurshid
  • Published in
    IEEE 30th International…
    2019
  • Computer Science
  • Deep Neural Networks (DNN) are increasingly used in a variety of applications, many of them with serious safety and security concerns. This paper describes DeepCheck, a new approach for validating DNNs based on core ideas from program analysis, specifically from symbolic execution. DeepCheck implements novel techniques for lightweight symbolic analysis of DNNs and applies them to address two challenging problems in DNN analysis: 1) identification of important input features and 2) leveraging… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 38 REFERENCES

    Safety Verification of Deep Neural Networks

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Concolic Testing for Deep Neural Networks

    VIEW 3 EXCERPTS

    DeepRoad: GAN-Based Metamorphic Testing and Input Validation Framework for Autonomous Driving Systems

    VIEW 1 EXCERPT

    Axiomatic Attribution for Deep Networks

    VIEW 1 EXCERPT

    DeepTest: Automated Testing of Deep-Neural-Network-Driven Autonomous Cars

    VIEW 2 EXCERPTS

    One Pixel Attack for Fooling Deep Neural Networks

    VIEW 1 EXCERPT

    Solving linear arithmetic with SAT-based model checking

    VIEW 1 EXCERPT