Posted: 02-06-2024

Location: U.S. Army Combat Capabilities Development Command-Aberdeen Proving Ground

Level: Graduate Student

General Topic: Computer Science  

Description of Research: Machine Learning has become an integral part of many domains (e.g., image analysis, networking protocols, network security, etc.), resulting in increased integration of ML into cyber defense tools. One way in which adversaries have responded is by perturbing inputs to cause misclassification to achieve their objective. This type of attack is known as adversarial machine learning (AML). Cybersecurity-related defenses to AML should strive to defend against unseen attacks and not require constant updating based on newly discovered attacks. Increasingly, supervised learning relies on a significant amount of labeled data to perform supervised learning. To avoid the requirements of a significant amount of labeled data, it is necessary to innovate self-supervised methodologies in a resource-constrained domain for network communications in the cyber domain. In the network/communications domain, machine learning-based classifiers are generally trained within a closed environment. Specifically, datasets used for training and evaluation are static and do not vary. Conversely, network environments are dynamic over time. Adversaries’ attacks become more sophisticated and change in response to defenders’ actions, requiring a defender to retrain a classifier to reflect the new attacks in the intended environment for deployment. This research seeks to address key research questions, such as: 

  • How do we design ML for cyber classifiers using a limited amount of data in a resource-constrained environment?  
  • How do we innovate network communication classifiers that are adversarial resilient?
  • Find a Volunteering Opportunity

    Visit our Program Volunteers page for a tool to find the best opportunity for you.



    The eCYBERMISSION Mini-Grant is intended to support teachers/program leaders as they implement eCYBERMISSION with their teams. Educators (formal and informal) of students in grades 6-9 are encouraged to apply. Special consideration is given to Title 1 schools and to those with underserved/ under-represented populations.


  • Latest AEOP News

    Request for Proposal: New evaluation partner needed to support the development and management of program evaluations, April 3, 2024

    See Story