计算机科学 ›› 2021, Vol. 48 ›› Issue (5): 301-307.doi: 10.11896/jsjkx.200800174
周天阳, 曾子懿, 臧艺超, 王清贤
ZHOU Tian-yang, ZENG Zi-yi, ZANG Yi-chao, WANG Qing-xian
摘要: 自动化渗透测试通过将人工找寻可能攻击路径的过程自动化,可大幅降低渗透测试的成本。现有方法主要利用单一Agent执行攻击任务,导致攻击动作执行耗时长,渗透效率不高;若考虑多个Agent协同攻击,由于每个Agent的局部状态有多个维度,总的规划问题的状态空间会呈指数级增长。针对上述问题,提出了基于多Agent联合决策的队组协同攻击规划方法。该方法首先将多Agent协同攻击路径规划问题转化为联合决策约束下的攻击目标分配问题,建立多Agent集中决策模式;然后以CDSO-CAP为模型基础,利用联合决策矢量矩阵JDVM计算渗透攻击奖励,并采用贪婪策略搜索多Agent的最优攻击目标。实验结果表明,与单Agent规划方法相比,该方法的收敛性相近,但执行轮次更短,更适合在多目标网络场景内进行快速攻击规划。
中图分类号:
[1]OBES J L,SARRAUTE C,RICHARTE G.Attack planning inthe real world[J].arXiv:1306.4044,2013. [2]SARRAUTE C,RICHARTE G,OBESJ L.AN algorithm to find optimal attack paths in nondeterministic scenarios[C]//Proceedings of th 4th ACM Workshop on Security and Artificial Intelligence.ACM,2011:71-80. [3]SARRAUTE C,BUFFET O,HOFFMANN J.POMDPs makebetter hackers:Accounting for uncertainty in penetration testing[C]//Twenty-Sixth AAAI Conference on Artificail Intelligence.2012. [4]SHMARYAHU D,SHANI G,HOFFMANN J,et al.Partially observable contingent planning for penetration testing[C]//IWAISe:First International Workshop on Artificail Intellijgence in Security.2017,33. [5]MCLENNAN A .The expected number of Nash equilibria of a normal form game[J].Econometrica,2005,73(1):141-174. [6]BOUTILIER C.Planning,learning and coordination in multi-agent decision processes[C]//Proceedings of the 6th conference on Theoretical aspects of rationality and knowledge.Morgan Kaufmann Publishers Inc.1996:195-210. [7]MUSLINER D J,DURFEE E H,WU J,et al.Coordinated Plan Management Using Multiagent MDPs[C]//AAAI Spring Symposium:DIstributed Plan and Schedule Management.2006:73-70. [8]ONTANON S,BURO M.Adversarial hierachical-task network planning for complex real-time games[C]//Twenty-Fourth International Joint Conference on Artificail Intelligence.2015. [9]SARRAUTE C,BUFFET O,HOFFMANN J.Penetration Testing==POMDP Solving?[J].arXiv:1306.4714,2013. [10]KOTENKO I.Agent-based modeling and simulation of cyber-warfare between malefactors and security agents in Internet[C]//the 19th European Simulation Multiconference “Simulation in wider Europe”.2005. [11]ROTH M,SIMMONS R,VELOSO M.What to communicate? Execution-time decision in multi-agent POMDPs[M]//Distributed Autonomous Robotic Systems 7.Springer,Tokyo,2006:177-186. [12]ZHANG C,LESSER V.Coordinated multi-agent reinforcement learning in networked distributed POMDPs[C]//Proceedings of the 25th AAAI Conference on Artificial Intelligence.San Francisco,America,2011:764-770. [13]ZHOU T Y,ZANG Y C,ZHU J H,et al.NIG-AP:a new me-thod for automated penetration testing[J].Frontiers of Information Technology & Electronic Engineering,2019,20(9):1277-1298. |
[1] | 冷典典, 杜鹏, 陈建廷, 向阳. 面向自动化集装箱码头的AGV行驶时间估计 Automated Container Terminal Oriented Travel Time Estimation of AGV 计算机科学, 2022, 49(9): 208-214. https://doi.org/10.11896/jsjkx.210700028 |
[2] | 熊丽琴, 曹雷, 赖俊, 陈希亮. 基于值分解的多智能体深度强化学习综述 Overview of Multi-agent Deep Reinforcement Learning Based on Value Factorization 计算机科学, 2022, 49(9): 172-182. https://doi.org/10.11896/jsjkx.210800112 |
[3] | 史殿习, 赵琛然, 张耀文, 杨绍武, 张拥军. 基于多智能体强化学习的端到端合作的自适应奖励方法 Adaptive Reward Method for End-to-End Cooperation Based on Multi-agent Reinforcement Learning 计算机科学, 2022, 49(8): 247-256. https://doi.org/10.11896/jsjkx.210700100 |
[4] | 王岩松, 秦云川, 蔡宇辉, 李肯立. 一种基于UIA接口的RPA系统设计方法 Design and Implementation of RPA System Based on UIA Interface 计算机科学, 2022, 49(8): 225-229. https://doi.org/10.11896/jsjkx.211100046 |
[5] | 高文龙, 周天阳, 朱俊虎, 赵子恒. 基于双向蚁群算法的网络攻击路径发现方法 Network Attack Path Discovery Method Based on Bidirectional Ant Colony Algorithm 计算机科学, 2022, 49(6A): 516-522. https://doi.org/10.11896/jsjkx.210500072 |
[6] | 张明新. 面向超大规模社会系统仿真的概念模型 Conceptual Model for Large-scale Social Simulation 计算机科学, 2022, 49(4): 16-24. https://doi.org/10.11896/jsjkx.210900136 |
[7] | 周仕承, 刘京菊, 钟晓峰, 卢灿举. 基于深度强化学习的智能化渗透测试路径发现 Intelligent Penetration Testing Path Discovery Based on Deep Reinforcement Learning 计算机科学, 2021, 48(7): 40-46. https://doi.org/10.11896/jsjkx.210400057 |
[8] | 黄双芹, 刘英博, 黄向生. 模型驱动开发工具的自动化测试技术研究 Research on Automatic Testing Technology of Model Driven Development Tools 计算机科学, 2021, 48(6A): 568-571. https://doi.org/10.11896/jsjkx.201000139 |
[9] | 曹浩, 郭绍忠, 刘聃, 许瑾晨. 面向64位RISC-V的基础数学库自动化移植 Automatic Porting of Basic Mathematics Library for 64-bit RISC-V 计算机科学, 2021, 48(6): 41-47. https://doi.org/10.11896/jsjkx.201200058 |
[10] | 高枫越, 王琰, 朱铁兰. 有适应力的分布式状态估计方法 Resilient Distributed State Estimation Algorithm 计算机科学, 2021, 48(5): 308-312. https://doi.org/10.11896/jsjkx.200300117 |
[11] | 左剑凯, 吴杰宏, 陈嘉彤, 刘泽源, 李忠智. 异构无人机编队防御及评估策略研究 Study on Heterogeneous UAV Formation Defense and Evaluation Strategy 计算机科学, 2021, 48(2): 55-63. https://doi.org/10.11896/jsjkx.191100053 |
[12] | 刘芳, 洪玫, 王潇, 郭丹, 杨正卉, 黄小丹. 面向Java的Randoop自动化单元测试生成工具性能分析 Performance Analysis of Randoop Automated Unit Test Generation Tool for Java 计算机科学, 2020, 47(9): 24-30. https://doi.org/10.11896/jsjkx.200200116 |
[13] | 罗云芳, 唐承娥, 韦军. 基于粗糙规则的脉冲神经膜系统计算能力的研究 Computing Ability of Spiking Neural P System Based on Rough Rules 计算机科学, 2020, 47(6A): 626-630. https://doi.org/10.11896/JsJkx.190500120 |
[14] | 孟繁祎, 王莹, 于海, 朱志良. 复杂软件系统的重构技术:现状、问题与展望 Refactoring of Complex Software Systems Research:PresentProblem and Prospect 计算机科学, 2020, 47(12): 1-10. https://doi.org/10.11896/jsjkx.200800067 |
[15] | 柴锐, 薛凡, 曾建潮, 秦品乐. 一种医学肾动态显像自动化定量评估方法 Automatic Quantitative Evaluation Approach for Medical Renal Dynamic Imaging 计算机科学, 2019, 46(8): 321-326. https://doi.org/10.11896/j.issn.1002-137X.2019.08.053 |
|