Dynamic Cyberattack Simulation: Integrating Improved Deep Reinforcement Learning with the MITRE-ATT&CK Framework

Author:

Oh Sang Ho1,Kim Jeongyoon2,Park Jongyoul2ORCID

Affiliation:

1. Department of Computer Engineering and Artificial Intelligence, Pukyong National University, Busan 48513, Republic of Korea

2. Department of Applied Artificial Intelligence, Seoul National University of Science and Technology, Seoul 01811, Republic of Korea

Abstract

As cyberattacks become increasingly sophisticated and frequent, it is crucial to develop robust cybersecurity measures that can withstand adversarial attacks. Adversarial simulation is an effective technique for evaluating the security of systems against various types of cyber threats. However, traditional adversarial simulation methods may not capture the complexity and unpredictability of real-world cyberattacks. In this paper, we propose the improved deep reinforcement learning (DRL) algorithm to enhance adversarial attack simulation for cybersecurity with real-world scenarios from MITRE-ATT&CK. We first describe the challenges of traditional adversarial simulation and the potential benefits of using DRL. We then present an improved DRL-based simulation framework that can realistically simulate complex and dynamic cyberattacks. We evaluate the proposed DRL framework using a cyberattack scenario and demonstrate its effectiveness by comparing it with existing DRL algorithms. Overall, our results suggest that DRL has significant potential for enhancing adversarial simulation for cybersecurity in real-world environments. This paper contributes to developing more robust and effective cybersecurity measures that can adapt to the evolving threat landscape of the digital world.

Funder

SeoulTech

Publisher

MDPI AG

Reference44 articles.

1. Harnessing artificial intelligence capabilities to improve cybersecurity;Zeadally;IEEE Access,2020

2. Machine learning methods for cyber security intrusion detection: Datasets and comparative study;Kilincer;Comput. Netw.,2021

3. Deep learning for cyber security intrusion detection: Approaches, datasets, and comparative study;Ferrag;J. Inf. Secur. Appl.,2020

4. Camlpad: Cybersecurity autonomous machine learning platform for anomaly detection;Hariharan;Advances in Information and Communication: Proceedings of the 2020 Future of Information and Communication Conference (FICC), San Francisco, CA, CA, USA, 5–6 March 2020,2020

5. Deep learning and machine learning based anomaly detection in internet of things environments;Calhan;J. Fac. Eng. Archit. Gazi Univ.,2022

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3