Conference: Resilient Cooperative AI/ML in Compromised Environments

Cyber Pack Ventures, Inc., Ellicott City MD

Investigators

Abstract

The workshop series, Computational Cybersecurity in Compromised Environments (C3E), explores the impact of recent technological advances and innovations in cyber-physical systems that must operate in environments that might themselves be hazardous and contested spaces. This project is part of the C3E workshop series, known as the “Challenge Problems”, where students as members of teams, mentored by researchers from government, industry and academia, conduct novel case studies, motivated by the themes of the workshop. These special studies are structured to provide insights into the challenges of adaption and adoption of new technologies and their implications for system engineering, operations, behaviors, and what is knowable. Artificial Intelligence / Machine Learning (AI/ML) technologies are increasingly part of a broad range of systems and a current subject of great interest and potentially considerable benefit to society yet, these new systems are raising profound concerns for system safety, dependability, security, resilience, and the evolving relationship between humans and technology. The Challenge Problems for this workshop focus on the role of AI/ML in critical infrastructure systems operating in these compromised environments where decision support involves human teaming and where resilience is seen through the lens of system risk and risk mitigation. Topics for the challenge problems are prepared largely by students in cooperation with their faculty advisors and mentors. The topics are reviewed by a subcommittee of the workshop program and organizing committees. Those teams whose topics are selected present the results of their year-long studies at the opening plenary session at the C3E workshop the following year. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.