通信与信息技术Issue(3)：6-10,33,6.
基于强化学习的自适应网络威胁缓解
Adaptive network threat mitigation based on reinforcement learning
齐分岭 1刘智磊 1张永军 1许延峰 1石成豪2
作者信息
- 1. 中国人民解放军66389部队,山西太原 030031
- 2. 航天工程大学研究生院,北京 101400
- 折叠
摘要
Abstract
With the in-depth development of internet information technology,communication networks are constantly threatened by attacks and intrusions.A reinforcement learning algorithm is proposed for network adaptive threat mitigation.Based on the SDN frame-work,the use of reinforcement learning algorithms for network security management is studied.Based on the D3QN algorithm and im-proved its structure,the improved D3QN deep reinforcement learning method is used to learn and mitigate APT attacks,achieving adaptive control of network threats.Finally,the experimental results were evaluated and the convergence results of the improved algorithm model were provided,verifying the availability and effectiveness of the reinforcement learning method for adaptive network threat mitigation.关键词
强化学习/SDN/改进D3QN算法/自适应网络威胁缓解Key words
Reinforcement learning/SDN/Improved D3QN algorithm/Adaptive network threat mitigation分类
信息技术与安全科学引用本文复制引用
齐分岭,刘智磊,张永军,许延峰,石成豪..基于强化学习的自适应网络威胁缓解[J].通信与信息技术,2024,(3):6-10,33,6.
