Research Interests

My research focuses on developing data-driven modeling frameworks that integrate machine learning with mathematical and mechanistic modeling to better understand, explain, and predict complex biological and environmental systems.

A central theme of my work is modeling systems whose dynamics change over time. I develop methods for discovering governing equations from data, identifying and forecasting time-varying processes, detecting early warning signals of critical transitions, and investigating causal relationships between system components and external drivers. By combining mechanistic structure with data-driven methods, I aim to move beyond prediction toward a deeper understanding of the processes responsible for observed system behavior.

My research spans a range of applications, including infectious disease dynamics, ecological systems, cellular processes, and greenhouse gas forecasting. I am particularly interested in understanding how internal mechanisms, environmental drivers, and external interventions jointly shape system behavior, and in distinguishing predictive associations from relationships that have a plausible mechanistic or causal basis.

Another major direction of my work is the development of predictive and causal frameworks for early warning and decision-making. This includes forecasting emerging changes in biological and environmental systems, evaluating responses to policy or environmental changes, identifying potential drivers of system transitions, and designing models that can help inform interventions.

Overall, my goal is to develop computational approaches that combine the interpretability and causal insight of mechanistic models with the predictive flexibility of machine learning. Through this integration, I aim to improve forecasting accuracy, strengthen model interpretability, uncover mechanisms and drivers of change, and deepen our understanding of complex, time-dependent systems.