Smart Contract Vulnerability Detection: From Pure Neural Network to Interpretable Graph Feature and Expert Pattern Fusion

Liu, Zhenguang; Qian, Peng; Wang, Xiang; Zhu, Lei; He, Qinming; Ji, Shouling

doi:10.24963/ijcai.2021/379

Smart Contract Vulnerability Detection: From Pure Neural Network to Interpretable Graph Feature and Expert Pattern Fusion

Zhenguang Liu, Peng Qian, Xiang Wang, Lei Zhu, Qinming He, Shouling Ji

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Main Track. Pages 2751-2759.

https://doi.org/10.24963/ijcai.2021/379

Keywords

Machine Learning: Knowledge Aided Learning
Machine Learning: Explainable/Interpretable Machine Learning
Multidisciplinary Topics and Applications: Security and Privacy

Abstract

Smart contracts hold digital coins worth billions of dollars, their security issues have drawn extensive attention in the past years. Towards smart contract vulnerability detection, conventional methods heavily rely on fixed expert rules, leading to low accuracy and poor scalability. Recent deep learning approaches alleviate this issue but fail to encode useful expert knowledge. In this paper, we explore combining deep learning with expert patterns in an explainable fashion. Specifically, we develop automatic tools to extract expert patterns from the source code. We then cast the code into a semantic graph to extract deep graph features. Thereafter, the global graph feature and local expert patterns are fused to cooperate and approach the final prediction, while yielding their interpretable weights. Experiments are conducted on all available smart contracts with source code in two platforms, Ethereum and VNT Chain. Empirically, our system significantly outperforms state-of-the-art methods. Our code is released.

Zhenguang Liu, Peng Qian, Xiang Wang, Lei Zhu, Qinming He, and Shouling Ji. 2021. Smart Contract Vulnerability Detection: From Pure Neural Network to Interpretable Graph Feature and Expert Pattern Fusion. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 2751–2759. https://doi.org/10.24963/ijcai.2021/379

@inproceedings{ijcai2021p379,
  title     = {Smart Contract Vulnerability Detection: From Pure Neural Network to Interpretable Graph Feature and Expert Pattern Fusion},
  author    = {Liu, Zhenguang and Qian, Peng and Wang, Xiang and Zhu, Lei and He, Qinming and Ji, Shouling},
  booktitle = {Proceedings of the Thirtieth International Joint Conference on
               Artificial Intelligence, {IJCAI-21}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Prof. Zhi-Hua Zhou},
  pages     = {2751--2759},
  year      = {2021},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2021/379},
  url       = {https://doi.org/10.24963/ijcai.2021/379},
}

TY  - CPAPER
TI  - Smart Contract Vulnerability Detection: From Pure Neural Network to Interpretable Graph Feature and Expert Pattern Fusion
AU  - Liu, Zhenguang
AU  - Qian, Peng
AU  - Wang, Xiang
AU  - Zhu, Lei
AU  - He, Qinming
AU  - Ji, Shouling
T2  - Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2021
SP  - 2751
EP  - 2759
DO  - 10.24963/ijcai.2021/379
UR  - https://doi.org/10.24963/ijcai.2021/379
ER  -