Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 2 of 2
Full-Text Articles in Entire DC Network
Generative Machine Learning For Cyber Security, James Halvorsen, Dr. Assefaw Gebremedhin
Generative Machine Learning For Cyber Security, James Halvorsen, Dr. Assefaw Gebremedhin
Military Cyber Affairs
Automated approaches to cyber security based on machine learning will be necessary to combat the next generation of cyber-attacks. Current machine learning tools, however, are difficult to develop and deploy due to issues such as data availability and high false positive rates. Generative models can help solve data-related issues by creating high quality synthetic data for training and testing. Furthermore, some generative architectures are multipurpose, and when used for tasks such as intrusion detection, can outperform existing classifier models. This paper demonstrates how the future of cyber security stands to benefit from continued research on generative models.
Evaluating Machine Learning Classifiers For Defensive Cyber Operations, Michael D. Rich, Robert F. Mills, Thomas E. Dube, Steven K. Rogers
Evaluating Machine Learning Classifiers For Defensive Cyber Operations, Michael D. Rich, Robert F. Mills, Thomas E. Dube, Steven K. Rogers
Military Cyber Affairs
Today’s defensive cyber sensors are dominated by signature-based analytical methods that require continuous maintenance and lack the ability to detect unknown threats. Anomaly detection offers the ability to detect unknown threats, but despite over 15 years of active research, the operationalization of anomaly detection and machine learning for Defensive Cyber Operations (DCO) is lagging. This article provides an introduction to machine learning concepts with a focus on the unique challenges to using machine learning for DCO. Traditional machine learning evaluation methods are challenged in favor of a value-focused evaluation method that incorporates evaluator-specific weights for classifier and sensitivity threshold selection …