Research Profile

Observability, inference, and decision-making for infrastructure resilience

I study how hidden states and risks in complex physical systems can be inferred from incomplete and heterogeneous observations, and how this information can support decision-making under uncertainty. My research combines machine learning, network science, geospatial data, and remote sensing to understand how local observations relate to system-level structure, behavior, and risk. I am particularly interested in climate, energy, and infrastructure systems, where critical states are often only partially observable and where disturbances can propagate across interconnected systems. My work explores methods for state estimation, risk characterization, and data-driven modeling that can improve our understanding and resilience of these systems.

Physical Systems →Sensing →Inference →Decision