Home : Research : Research Labs : ICSARL

Intelligent Cyber-Systems and Analytics Research Laboratory (ICSARL)

RESEARCH OVERVIEW

Our research directly supports the overarching ACI research initiatives of cyber autonomy, cyber resiliency for critical infrastructures, and cyber-physical systems. As part of our research agenda, we conduct basic and applied cyber research in the areas of:

  • Artificial intelligence and machine/deep learning
  • Data science, data mining and big data analytics
  • Computational science and statistical computing
  • Knowledge representation, discovery and network science
  • High-performance, distributed computing and parallel processing
RESEARCH FOCUS

Our research focuses on the design, development, experimentation and operationalization of computationally intelligent, large-scale, and distributed decision-support systems for autonomous cyber operations. To build, test and deploy autonomous cyber-systems that enable intelligent cybersecurity data analytics, our research team explores the science of information, computation, learning, and fusion as it pertains to the use of artificial intelligence for adaptive pattern discovery, learning, reasoning, perception, action and decision-making given heterogenous, disparate cyber data.

Our research aims to provide the Army Cyber Enterprise new capabilities to:

  • Shift emphasis from sensing to information awareness
  • Understand the underpinning of cyber autonomy
  • Relieve human cybersecurity analyst cognitive overload in dealing with the data deluge problem
  • Enhance human-machine interface in information processing

Our research has the potential to:

  • Cope with various complex disparate data/information types
  • Integrate a diversity of unique reasoning and learning components collaborating simultaneously
  • Bridge correlational with causal discovery
  • Determine solutions or obstructions to local-to-global data fusion problems
  • Mechanize reasoning/learning and computing in the same computational environment
  • Yield provably efficient procedures to enable or facilitate cyber data analytics
  • Deal with high-dimensional and massive cyber datasets with provably guaranteed performance

Our current research efforts include:

  • "Cyber Pattern Learning for Autonomous Network Vulnerability Assessment and Decision Support at Scale"
  • "Feature Selection for High-Performance Computing Enabled Deep Learning in the Cybersecurity Domain"
  • "Intelligent Systems Design for Malware Classification under Adversarial Conditions"
  • "Cyber Resiliency for Critical Infrastructures: Real-Time, Large-Scale Anomaly Detection in Smart Grids"
  • "Deep Reinforcement Learning for Sequential Resource Allocation Decisions Under Uncertainty: An Application in Cyber Overwatch Hunt Teams"
RESEARCH PARTNERS

U.S. Army Research Laboratory (ARL)

U.S. Army Engineer Research and Development Center (ERDC)

U.S. Army Cyber Command (ARCYBER)

Defense Advanced Research Projects Agency (DARPA)

Network Enterprise Technology Command (NETCOM)

National Security Agency (NSA)

U.S. Air Force Research Laboratory (AFRL)

Johns Hopkins University Applied Physics Laboratory (JHUAPL)

U.S. Army Futures Command (AFC)