Applying reinforcement learning in automated penetration testing
DOI:
https://doi.org/10.54654/isj.v3i17.876Keywords:
Penetration testing, reinforcement learning, information secutityTóm tắt
Abstract— Facing increasingly diverse and frequent information security threats today, penetration testing is a security assessment method for information systems that organizations prioritize. Pentesters usually perform penetration testing manually and can detect critical bugs and information security issues. However, this method requires much work and requires pentesters to have high levels of practical experience and qualifications. One of the current research directions that has been interested recently is methods to support automated penetration testing. Several research groups have used attack graph analysis techniques and reinforcement learning algorithms worldwide to make automated pentesting tools. This paper proposes a model based on a reinforcement learning algorithm and parameter optimization method for this model in automated pentesting problems. To evaluate the proposed model, we utilize the data set based on the method used by other research groups. We also assess the self-built dataset on real environments with vulnerabilities. The experimental results show that the proposed method gives better assessments than other methods.
Downloads
References
X. Y. B. T. B. C. M. J. Aileen G. Bacudio, “An Overview of Penetration Testing,” International Journal of Network Security And Its Applications (IJNSA), pp. Vol.3, No.6, November 2011.
M. E. Farmeena Khan, “A Comparative Study of White Box, Black Box and Grey Box Testing Techniques,” in International Journal of Advanced Computer Science and Applications, June 2012.
C. B. S. a. S. S. Vivek Shandilya, “Use of Attack Graphs in Security Systems,” Journal of Computer Networks and Communications, 2014.
R. L. a. K. Ingols, “An Annotated Review of Past Papers on Attack Graphs,” in Lincoln Laboratory MASSACHUSETTS INSTITUTE OF TECHNOLOGY, 2005.
L. P. S. Cynthia Phillips, “A Graph-Based System for Network Vulnerability Analysis,” in Proceedings of the 1998 workshop on New security paradigms, 1998.
J. H. S. J. e. a. O. Sheyner, “Automated generation and analysis of attack graphs,” in (IJACSA) Proceedings of the IEEE Symposium on Security and Privacy, Berkeley, CA USA, May 2022.
N. M. Y. Z. a. H. T. Mehdi Yousefi, “A Reinforcement Learning Approach for Attack Graph Analysis,” in IEEE International Conference on Trust Security and Privacy in Computing and Communications (TrustCom), 2018.
P. Đ. K. N. Đ. V. N. V. H. Nguyễn Mạnh Thiên, “Phát triển Framework ứng dụng AI hỗ trợ tự động khai thác lỗ hổng bảo mật,” Journal of Science and Technology on Information Security, vol. 1, no. 13, pp. 80-92, 2022.
“OpenVas,” [Online]. Available: https://openvas.org. [Accessed 20 04 2022].
S. G. A. W. A. Xinming Ou, “MulVAL: A Logic-based Network Security Analyzer,” in USENIX Security Sympo, 31 July 2005.
“Datalog,” [Online]. Available: 2022]. Available: https://en.wikipedia.org/wiki/Datalog. [Accessed 20 6 2022].
S. G. A. A. Xinming Ou, “MulVAL: A Logic-based Network Security Analyzer,” in USENIX Security Symposium, 31 July 2005.
B. J. K. W. a. F. D. Marcin Szpyrka, “Telecommunications Networks Risk Assessment with Bayesian Networks,” in Computer Information Systems and Industrial Management Proceedings of the 12th IFIP TC8 International Conference CISIM, 2013.
R. B. Y. T. Zhenguo Hu, “Automated Penetration Testing Using Deep Reinforcement Learning,” in 2020 IEEE European Symposium on Security and Privacy Workshops (EuroS and PW), 2020.
“GNS3,” [Online]. Available: https://www.gns3.com. [Accessed 20 06 2022].
“Metasploit,” [Online]. Available: https://www.offensive security.com/metasploit-unleashed/pivoting/. [Accessed 20 06 2022].
“Metasploit Pivoting,” [Online]. Available: https://www.offensivesecurity.com/metasploitunleashed/pivoting/. [Accessed 22 06 2022].
“Pymetasploit,” [Online]. Available: https://github.com/DanMcInerney/pymetasploit3[Accessed 20 06 2022].
S. W. Y. C. R. Z. a. C. W. Ianping Zeng, “Survey of Attack Graph Analysis Methods from the Perspective of Data and Knowledge Processing,” Security and Communication Networks, vol. 2019, no. 2031063, 2019.
Downloads
Published
How to Cite
Issue
Section
License
Proposed Policy for Journals That Offer Open Access
Authors who publish with this journal agree to the following terms:
1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
Proposed Policy for Journals That Offer Delayed Open Access
Authors who publish with this journal agree to the following terms:
1. Authors retain copyright and grant the journal right of first publication, with the work [SPECIFY PERIOD OF TIME] after publication simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
