Welcome to the Home of the Numerical Intelligent Systems Laboratory

The Numerical Intelligent Systems Laboratory is a collaborative laboratory between Montana State University and the Johns Hopkins University, focusing on fundamental problems in artificial intelligence and machine learning from a numerical computation perspective. We are exploring problems in advanced knowledge representation, inference, and learning as it applies to system-level problems such as system monitoring and control, equipment health management, and bio-inspired design. Techniques explored include probabilistic and Bayesian methods, evolutionary methods, and particle-based methods. We are also interested in exploring the role of background knowledge in improving performance in learning and knowledge discovery.

Projects

• Augmented and Not-So-Naive Bayesian Models
• Semantic Interoperability of Diagnostic Systems
• Dynamic Bayesian Networks and Prognostics & Health Management (PHM)
• Asset-Specific Models and Distributional Smoothing
• Applications of Particle Swarm Optimization
• Ontology-Directed Data Mining
• Dimensionality Reduction for CBIR
• Machine Learning and Music