ABSTRACT
In recent years, enterprises have been targeted by advanced adversaries who leverage creative ways to infiltrate their systems and move laterally to gain access to critical data. One increasingly common evasive method is to hide the malicious activity behind a benign program by using tools that are already installed on user computers. These programs are usually part of the operating system distribution or another user-installed binary, therefore this type of attack is called “Living-Off-The-Land”. Detecting these attacks is challenging, as adversaries may not create malicious files on the victim computers and anti-virus scans fail to detect them.
We propose the design of an Active Learning framework called LOLAL for detecting Living-Off-the-Land attacks that iteratively selects a set of uncertain and anomalous samples for labeling by a human analyst. LOLAL is specifically designed to work well when a limited number of labeled samples are available for training machine learning models to detect attacks. We investigate methods to represent command-line text using word-embedding techniques, and design ensemble boosting classifiers to distinguish malicious and benign samples based on the embedding representation. We leverage a large, anonymized dataset collected by an endpoint security product and demonstrate that our ensemble classifiers achieve an average F1 score of 96% at classifying different attack classes. We show that our active learning method consistently improves the classifier performance, as more training data is labeled, and converges in less than 30 iterations when starting with a small number of labeled instances.
- Mohamad Mahmoud Al Rahhal, Yakoub Bazi, Haikel AlHichri, Naif Alajlan, Farid Melgani, and Ronald R Yager. 2016. Deep learning approach for active classification of electrocardiogram signals. Information Sciences 345(2016), 340–354.Google ScholarDigital Library
- M. Almgren and E. Jonsson. 2004. Using active learning in intrusion detection. In Proc. IEEE Computer Security Foundations Workshop. 88–98.Google Scholar
- Dana Angluin. 1988. Queries and concept learning. Machine learning 2, 4 (1988), 319–342.Google Scholar
- AppArmor. 2021. AppArmor: Linux kernel security module. https://apparmor.net/Google Scholar
- Ignacio Arnaldo, Kalyan Veeramachaneni, and Mei Lam. 2019. eX2: a framework for interactive anomaly detection.. In IUI Workshops.Google Scholar
- Les Atlas, David Cohn, and Richard Ladner. 1989. Training connectionist networks with queries and selective sampling. Advances in neural information processing systems 2 (1989), 566–573.Google Scholar
- Anaël Beaugnon, Pierre Chifflier, and Francis Bach. 2017. Ilab: An interactive labelling strategy for intrusion detection. In International Symposium on Research in Attacks, Intrusions, and Defenses. Springer, 120–140.Google ScholarCross Ref
- Battista Biggio, Blaine Nelson, and Pavel Laskov. 2012. Poisoning attacks against support vector machines. In ICML.Google Scholar
- Daniel Bohannon and Lee Holmes. 2017. Revoke-obfuscation: powershell obfuscation detection using science.Google Scholar
- Leo Breiman. 2001. Random forests. Machine learning 45, 1 (2001), 5–32.Google ScholarDigital Library
- Varin Chouvatut, Wattana Jindaluang, and Ekkarat Boonchieng. 2015. Training set size reduction in large dataset problems. In 2015 International Computer Science and Engineering Conference (ICSEC). IEEE, 1–5.Google ScholarCross Ref
- David A Cohn, Zoubin Ghahramani, and Michael I Jordan. 1996. Active learning with statistical models. Journal of artificial intelligence research 4 (1996), 129–145.Google ScholarDigital Library
- Crowdstrike. 2019. The Rise of “Living off the Land” Attacks | CrowdStrike. https://www.crowdstrike.com/blog/going-beyond-malware-the-rise-of-living-off-the-land-attacksGoogle Scholar
- Cytomic. 2019. Living-off-the-Land attacks: what are they and why should they worry you? | Cytomic. https://www.cytomicmodel.com/news/living-off-the-land-attacksGoogle Scholar
- Alessandra De Paola, Salvatore Gaglio, Giuseppe Lo Re, and Marco Morana. 2018. A hybrid system for malware detection on big data. In IEEE INFOCOM 2018-IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS). IEEE, 45–50.Google ScholarCross Ref
- Hervé Debar, Marc Dacier, Mehdi Nassehi, and Andreas Wespi. 1998. Fixed vs. variable-length patterns for detecting suspicious process behavior. In European Symposium on Research in Computer Security. Springer, 1–15.Google ScholarDigital Library
- Benoît Frénay and Michel Verleysen. 2013. Classification in the presence of label noise: a survey. IEEE transactions on neural networks and learning systems 25, 5(2013), 845–869.Google Scholar
- Nico Görnitz, Marius Kloft, Konrad Rieck, and Ulf Brefeld. 2009. Active learning for network intrusion detection. In Proceedings of the 2nd ACM workshop on Security and artificial intelligence. ACM, 47–54.Google ScholarDigital Library
- Nico Görnitz, Marius Kloft, Konrad Rieck, and Ulf Brefeld. 2013. Toward supervised anomaly detection. Journal of Artificial Intelligence Research 46 (2013), 235–262.Google ScholarDigital Library
- Tianyu Gu, Brendan Dolan-Gavitt, and Siddharth Garg. 2017. Badnets: Identifying vulnerabilities in the machine learning model supply chain. arXiv preprint arXiv:1708.06733(2017).Google Scholar
- M. Jagielski, A. Oprea, B. Biggio, C. Liu, C. Nita-Rotaru, and B. Li. 2018. Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning. In 2018 IEEE Symposium on Security and Privacy (SP). 19–35. https://doi.org/10.1109/SP.2018.00057Google Scholar
- Thorsten Joachims. 1998. Text categorization with support vector machines: Learning with many relevant features. In European conference on machine learning. Springer, 137–142.Google ScholarDigital Library
- Armand Joulin, Edouard Grave, Piotr Bojanowski, and Tomas Mikolov. 2016. Bag of tricks for efficient text classification. arXiv preprint arXiv:1607.01759(2016).Google Scholar
- David D Lewis and William A Gale. 1994. A sequential algorithm for training text classifiers. In SIGIR’94. Springer, 3–12.Google Scholar
- Tong Luo, Kurt Kramer, Dmitry B Goldgof, Lawrence O Hall, Scott Samson, Andrew Remsen, Thomas Hopkins, and David Cohn. 2005. Active learning to recognize multiple types of plankton.Journal of Machine Learning Research 6, 4 (2005).Google Scholar
- Carla Marceau. 2001. Characterizing the behavior of a program using multiple-length n-grams. In Proceedings of the 2000 workshop on New security paradigms. 101–110.Google ScholarDigital Library
- Tren Micro. 2021. Tracking, Detecting, and Thwarting PowerShell-based Malware and Attacks - Security News. https://www.trendmicro.com/vinfo/hk-en/security/news/cybercrime-and-digital-threats/tracking-detecting-and-thwarting-powershell-based-malware-and-attacksGoogle Scholar
- Microsoft. 2018. Out of sight but not invisible: Defeating fileless malware with behavior monitoring, AMSI, and next-gen AV - Microsoft Security. https://www.microsoft.com/security/blog/2018/09/27/out-of-sight-but-not-invisible-defeating-fileless-malware-with-behavior-monitoring-amsi-and-next-gen-avGoogle Scholar
- Microsoft. 2021. Microsoft Defender for Endpoint | Microsoft Security. https://www.microsoft.com/en-us/security/business/threat-protection/endpoint-defenderGoogle Scholar
- Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg S Corrado, and Jeff Dean. 2013. Distributed representations of words and phrases and their compositionality. In Advances in neural information processing systems. 3111–3119.Google Scholar
- Brad Miller, Alex Kantchelian, Sadia Afroz, Rekha Bachwani, Edwin Dauber, Ling Huang, Michael Carl Tschantz, Anthony D Joseph, and J Doug Tygar. 2014. Adversarial active learning. In Proceedings of the 2014 Workshop on Artificial Intelligent and Security Workshop. ACM, 3–14.Google ScholarDigital Library
- Palo Alto Networks. 2020. What Are Fileless Malware Attacks and “Living Off the Land”? Unit 42 Explains. https://www.paloaltonetworks.com/cyberpedia/what-are-fileless-malware-attacksGoogle Scholar
- D. Pelleg and A. Moore. 2004. Active Learning for Anomaly and Rare-Category Detection. In Proc. Advances in Neural Information Processing Systems. 1073–1080.Google Scholar
- GTFOBins Project. 2021. Living Off The Land Binaries for UNIX. https://gtfobins.github.io/.Google Scholar
- LOLBAS Project. 2021. Living Off The Land Binaries and Scripts (and also Libraries). https://lolbas-project.github.io/.Google Scholar
- Shubham Rai. 2020. Behavioral Threat Detection: detecting Living of Land Techniques. Master’s thesis. University of Twente.Google Scholar
- Juan Ramos 2003. Using tf-idf to determine word relevance in document queries. In Proceedings of the first instructional conference on machine learning, Vol. 242. Piscataway, NJ, 133–142.Google Scholar
- Korlakai Vinayak Rashmi and Ran Gilad-Bachrach. 2015. DART: Dropouts meet Multiple Additive Regression Trees.. In AISTATS. 489–497.Google Scholar
- N Roy and A McCallum. 2001. Toward optimal active learning through sampling estimation of error reduction. Int. Conf. on Machine Learning.Google ScholarDigital Library
- Amir Rubin, Shay Kels, and Danny Hendler. 2019. Detecting Malicious PowerShell Scripts Using Contextual Embeddings. arXiv preprint arXiv:1905.09538(2019).Google Scholar
- Gili Rusak, Abdullah Al-Dujaili, and Una-May O’Reilly. 2018. Ast-based deep learning for detecting malicious powershell. In Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security. ACM, 2276–2278.Google ScholarDigital Library
- Andrew I Schein and Lyle H Ungar. 2007. Active learning for logistic regression: an evaluation. Machine Learning 68, 3 (2007), 235–265.Google ScholarDigital Library
- D Sculley. 2007. Online active learning methods for fast label-efficient spam filtering.. In CEAS, Vol. 7. 143.Google Scholar
- D Sculley, Matthew Eric Otey, Michael Pohl, Bridget Spitznagel, John Hainsworth, and Yunkai Zhou. 2011. Detecting adversarial advertisements in the wild. In Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 274–282.Google ScholarDigital Library
- Burr Settles. 2009. Active Learning Literature Survey. Computer Sciences Technical Report 1648. University of Wisconsin–Madison.Google Scholar
- H Sebastian Seung, Manfred Opper, and Haim Sompolinsky. 1992. Query by committee. In Proceedings of the fifth annual workshop on Computational learning theory. 287–294.Google ScholarDigital Library
- Md Amran Siddiqui, Jack W Stokes, Christian Seifert, Evan Argyle, Robert McCann, Joshua Neil, and Justin Carroll. 2019. Detecting Cyber Attacks Using Anomaly Detection with Explanations and Expert Feedback. In ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2872–2876.Google Scholar
- Robin Sommer and Vern Paxson. 2010. Outside the Closed World: On Using Machine Learning for Network Intrusion Detection. In 2010 IEEE Symposium on Security and Privacy. 305–316. https://doi.org/10.1109/SP.2010.25Google ScholarDigital Library
- Ryan Stewart. 2019. Let’s dig deeper on how cybercriminals use ‘Living off the land’ attack tactics | Cyware Hacker News. Cyware (Mar 2019). https://cyware.com/news/lets-dig-deeper-on-how-cybercriminals-use-living-off-the-land-attack-tactics-cac5c132Google Scholar
- Jack W Stokes, Ashish Kapoor, and Debajyoti Ray. 2016. Asking for a second opinion: Re-querying of noisy multi-class labels. In 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2329–2333.Google ScholarDigital Library
- Jack W Stokes, John Platt, Joseph Kravis, and Michael Shilman. 2008. Aladin: Active learning of anomalies to detect intrusions. (2008).Google Scholar
- Symantec. 2017. Living off the land and fileless attack techniques. Technical Report.Google Scholar
- Symantec. 2021. PowerShell threats surge - Symantec Enterprise. https://community.broadcom.com/symantecenterprise/communities/community-home/librarydocuments/viewdocument?DocumentKey=cbd24b89-1022-4fe8-800d-a362f3d4cf06&CommunityKey=1ecf5f55-9545-44d6-b0f4-4e4a7f5f5e68&tab=librarydocumentsGoogle Scholar
- Jorge L Guerra Torres, Carlos A Catania, and Eduardo Veas. 2019. Active learning approach to label network traffic datasets. Journal of information security and applications 49 (2019), 102388.Google ScholarDigital Library
- Devis Tuia, Frédéric Ratle, Fabio Pacifici, Mikhail F Kanevski, and William J Emery. 2009. Active learning methods for remote sensing image classification. IEEE Transactions on Geoscience and Remote Sensing 47, 7 (2009), 2218–2232.Google ScholarCross Ref
- Kalyan Veeramachaneni, Ignacio Arnaldo, Vamsi Korrapati, Constantinos Bassias, and Ke Li. 2016. AI2303 2: training a big data machine to defend. In 2016 IEEE 2nd International Conference on Big Data Security on Cloud (BigDataSecurity), IEEE International Conference on High Performance and Smart Computing (HPSC), and IEEE International Conference on Intelligent Data and Security (IDS). IEEE, 49–54.Google Scholar
- Venafi. 2020. Beware of Cyber Attackers “Living off the Land” | Venafi. https://www.venafi.com/blog/beware-cyber-attackers-living-landGoogle Scholar
- Qi Wang, Wajih Ul Hassan, Ding Li, Kangkook Jee, Xiao Yu, Kexuan Zou, Junghwan Rhee, Zhengzhang Chen, Wei Cheng, C Gunter, 2020. You are what you do: Hunting stealthy malware via data provenance analysis. In Proc. of the Symposium on Network and Distributed System Security (NDSS).Google ScholarCross Ref
- Colin Whittaker, Brian Ryner, and Marria Nazif. 2010. Large-scale automatic classification of phishing pages. (2010).Google Scholar
- Muhammd Mudassar Yamin and Basel Katt. 2018. Detecting Malicious Windows Commands Using Natural Language Processing Techniques. In International Conference on Security for Information Technology and Communications. Springer, 157–169.Google Scholar
- Living-Off-The-Land Command Detection Using Active Learning
Recommendations
Unraveled — A semi-synthetic dataset for Advanced Persistent Threats
AbstractU n r a v e l e d is a novel cybersecurity dataset capturing Advanced Persistent Threat (APT) attacks not available in the public domain. Existing cybersecurity datasets lack coherent information about sophisticated and persistent cyber-attack ...
Using Supervised Learning to Detect Command and Control Attacks in IoT
The rapid proliferation of internet of things (IoT) devices has ushered in a new era of technological development. However, this growth has also exposed these devices to various cybersecurity risks, including command and control (C&C) attacks. C&C ...
EarlyCrow: Detecting APT Malware Command and Control over HTTP(S) Using Contextual Summaries
Information SecurityAbstractAdvanced Persistent Threats (APTs) are among the most sophisticated threats facing critical organizations worldwide. APTs employ specific tactics, techniques, and procedures (TTPs) which make them difficult to detect in comparison to frequent and ...
Comments