A-4: Advanced Research in Cyber Systems

AI Assurance

Capabilities in: Depth Completion on Autonomous Vehicles using sparse coding, Detecting Deepfakes and other generated media, Evaluating the Adversarial Robustness of Energy-Based Models, Defense against Model Inversion Attacks, and Robust Audio Classification using Multi-layer Neural Networks.

Assured Computing/Computational Integrity

Capabilities include: Quantum Key Distribution, Cryptographically Robust and Assured AI/ML, Cryptographic Structure of Optical Physical Unclonable Functions, and Authenticating Internet Routing Using Zero-knowledge Proofs.

Secure Systems

Secure System Composition and Type Checking using Cryptographic Proofs Simultaneous Classification of Malware, Malware Families, and Novel Malware General-Purpose Unsupervised Cyber Anomaly Detection Electrical Grid Anomaly Detection in Smart Power Grids Evaluate Network Resiliency via Adversarial Abstraction Accelerated Network Environments for AI Cyber Arms Races

Lead DOE Cyberfire Incident Response Training Program

The DOE Cyberfire Incident Response Training Program takes the lead in equipping professionals with the skills necessary to handle cybersecurity incidents effectively. This program focuses on empowering individuals to respond efficiently to cyber threats within the Department of Energy.

Infrastructure Support

We develop, administer, maintain and support highly customized data systems on multiple networks for the Nuclear Emergency Support Team, Intelligence community, and related programs.

Data Sets