Global Journal of Computer Science and Technology, D: Neural & Artificial Intelligence, Volume 23 Issue 2

[3] Supervised Learning 47,398 Deep Learning ReVulDL is a deep learning-based two- phase smart contract debugger for re- entrancy vulnerability. It integrates the vulnerability detection and localization into a unified debugging pipeline. [4] Semi-Supervised Learning 20,829 BERT ASSBert is a model that uses active and semi-supervised learning with BERT for smart contract vulnerability detection. It aims to improve the accuracy and scalability of vulnerability detection by combining deep learning with expert patterns in an explainable fashion. [5] Reinforcement Learning Not provided Reinforcement Learning and Fuzzing Vulnerability-guided fuzzer based on reinforcement learning, namely RLF, for generating vulnerable transaction seque- nces to detect sophisticated vulnerabilities in smart contracts. The experimental results demon-strate that RLF outperforms state-of- the-art vulnerability-detection tools. [6] Supervised Learning 40.932 GNN and Expert Knowledge The use of graph neural networks and expert knowledge for smart contract vulnerability detection. Empirical results show significant accuracy improvements over state-of-the-art methods on three types of vulnerabilities. [7] Supervised Learning 70,000 Machine Learning SmartMixModel is a machine learning- based vulnerability detection model for Solidity smart contracts. It considers an expanded feature space covering both the source- and byte codes of the Solidity smart contracts, and achieves improved detection performance compared to state of the art models. v. C onclusion In conclusion, there exists a compelling need for continued research into the application of artificial intelligence (AI) in the detection of flaws within smart contracts. This research seeks to offer invaluable insights through the comparative evaluation of existing AI-based algorithms for smart contract fault detection, shedding light on the efficacy of various AI approaches. While the potential of combining AI with formal methods has been acknowledged, there remains untapped potential in need of exploration. Future research endeavors in the realm of smart contracts will pivot towards the development of AI powered detection tools capable of addressing security breaches associated with smart contracts while handling large datasets efficiently and effectively. Additionally, attention must be directed towards the utilization of SSL (Semi-Supervised Learning) and RL (Reinforcement Learning) to potentially overcome the limitations of SL (Supervised Learning). A comprehensive investigation into smart contract flaw detection using AI is imperative, serving as a foundational reference and a wellspring of inspiration for forthcoming research. Ultimately, the integration of AI with formal techniques holds the promise of substantially enhancing the security of smart contracts, ensuring their reliability and robustness in blockchain-based applications R eferences R éférences R eferencias 1. X. Yu, H. Zhao, B. Hou, Z. Ying and B. Wu, "DeeSCVHunter: A Deep Learning-Based Framework for Smart Contract Vulnerability Detection," 2021 International Joint Conference on Neural Networks (IJCNN), Shenzhen, China, 2021, pp. 1-8, doi:10. 1109/IJCNN52387.2021.9534324. 2. Gupta, R.; Patel, M.M.; Shukla, A.; Tanwar, S. Deep learning-based malicious smart contract detection scheme for internet of things environment. Comput. Electr. Eng. 2022, 97, 107583. 3. Zhuo Zhang, Yan Lei, Meng Yan, Yue Yu, Jiachi Chen, Shangwen Wang, and Xiaoguang Mao. 2023. Reentrancy Vulnerability Detection and Localization: A Deep Learning Based Two-phase Approach. In Proceedings of the 37th IEEE/ACM International © 2023 Global Journals Global Journal of Computer Science and Technology Volume XXIII Issue II Version I 65 ( )D Year 2023 Strengthening Smart Contracts: An AI-Driven Security Exploration